”.35.. s 1.... . . ....x.r...........muvpv. . gawk? .. . . .. .. (F... ..:. r2... .1 18m.»- uu... 1.5.55“. .. i. .517. 3" z. . v .. -‘ I .f . . an: 1......“ 4.4M”? . n..- ....?¢M._.m..m w... mu... mun... .n.fl..n..h... 13?»... .m... 3:... . . . . Inifih...’ . b» ..a. , ..:? ..l .mfi.J....... .1 : . . .rr. .1 EN. . . a- . ...........u.....fi..v.... . . .. : $5.53». - . ~ r, .15. . ~ ..:“. iv. . _ _ rd... in 703.1. m . I .r. 5.... o .. . J ”hm... 5.6.1.1. r . ..mm........w $1.... . .mérw.3.1.m -. ”$3 1. lulu“ 1 5 - . 5?... a .I i .3- .31, ‘3. fl . o . 5. .. ...EflYS u. ..:. «1..le ...:. #5.. ..: a... .. 1:.ofi: :5. 0" u}. n‘ c ‘0 GA, 4‘ .1 ..: .. ..: . 51...... .293. ‘ 2 O ”-3. 2%.: .II ru.:VItv1‘$-§ fl ”Waning. Q vIo flat} I v ..: *3 :3 32 F: S. ‘ . z. 331,»... ...:! .. {v.1 3.3.23.1. 2. r. 1 it-.. s. .. 35.2? 1.2.4.»... an". a... “aquatic. 3.11....» “hula“: . o i. .. I. e. 1 m... visa 1.. #5”. ._.. . .u.........5d: . .1 .4 -6 . ...mfiwfififluibz .. ...n... .2. a. ..3 “(LET M Wlmur. stdnfilfi. 3| 7 tv I . x I. :0: A 1.... .1 . u o . .2; A.N\.c!. . 1“. t. .1 1< I. «up...» a... #fluuyééfizmm ..m “undenhvnuflqu «.1. Khaki}. . . - $4M,“ Rum... «lad... %1F .1. . .. .vs. ,I\ .11. 53-! .... ..wx_.fi...1.uvn...&... 3.. .. 1.239-.. . ...WL Wham. .....Wh...vh.2¢.4 I.“ .5131 Eitlts. . C... I ."ltTni, Ix. I x. .73.}: 0 . . . «1:25.... any... ..4w:n.....o.... .. .fina. “an... .u... 31.3.1.1 2...... ‘1 .. .5 .1 2.. ..3 .. fibfiamgnum... . 5...: ...uhau...Wnum$W~Jmmwm§hwh2ww 1 L13! .9 ...A......~,...!.t1....$. .. : .1... u L... ..h.wm._...-2.« 1 .4 f. , . 33:. xi ‘. u ' . -I ‘d c l i . ...:. . 1‘?! c. ...:.IMTELu . am... .L5 1.. ii... I......il:-M:u 2!: MI? uxamxdkwfl .59.“. a. .. #kumod... 5.12. ...Cr I ~...I.Iaqr\»th.ir.xwl tnr‘..oxvtloorlv . . . . . . ..u_u...~pw...u......fv.u 11 .........,..51......ul..... ..$...1..! .. . . .1. , 23...... - . . -. . . l. :5... 3 .1.-- 21...! L322. .2 ...: to... . : 1.2.3.. 2; (”in T .., «5...... 1W»- AI 7‘ .01 n . v.3! I . ‘9 ll v.59. . ....cr..nd....1£.;.n.n..! . I'll av»: a 1 It‘ll- . 25 l . .. ..ufi . ..:...4. 2.... 91...»... 3m“... n.nh.o3wnwivh.1vfi ..Amms...5vfit!nufl.flm.}$§r . 4-.. .. .2 .. 5,33%. .mxktrazit kirk}: huntfioux... uh... abngwkifln? . .... 5.2.1.21? 711.11. 3... u I. 1.134...“ .05.. ..: v x. ...:... .v... I: . .. in... 2.1. {n}... .131... $41.2: 33.32.19. v . v 1.11.; 1725:...» u... ...ur ... 1. . ...- :? m“...- ...r .. It . I. 1.31.: .11... .1555. ..:.t 3121.32.11.45 (If... ..: I .10 . Inctoonvvu..1:|!1.|.1v..|v.nsfih.ln9vtz 1v. \ {Medus- r.....1..1~. ...:. ...rmmvriu I 1 2.. Juan... . .. .. 9... on..!......k~.....$.. a». . ., 1k. .33.... Wuhan... 4 “Marga?! o.. 5.1. . .r...111:.a......uu...mn.: . . P1 1?. 3.1.9.9.??? .4. s11. . .1..fi..!.d.....a.......n.....uh ..Xr .. ...b. v . . Vinita. . .E. .(.$.r........¢..,.u. . ..cmn 2r. H.» 5.23:3! .1 .... .» 1...» «3th: ..v . . 21:30.5” ...:»...qm V.....ah..5-. .t _ . . . .. . . 1 {zeiwrmfinw .5» ..:.h......«&.... urv "1 flue...» 1 8w... .fiffi fifihbv. . . . 1"”..9; 113.9 ...:IV 1... t. p . ...l 1...... .. 1. 5k. v .. h .. u ...n .211, . 4.. (that. . ylmm . . | l . l I‘Vl'oIVI‘u VlJ...‘ , . . .. . L .. 1 .. ,..o.u.....u..flfl.uww.mv.m$.§5 -i. fiat... ..P 2.35.... .» ...:. ~ 1. a... . . a. rv ...flflgifiandngaflwfimuf. 3.:qu35.§.. . .. . , . . ..2..:.._.n.... .Muuommu 24.“... it. y .1 .vlllo..v . . . A I12|an . «WU? nu. 4 ...u‘.. ...dwwnnwédu. .. . Jabkmnmw. ...m...1.....n.~. . . .. .. .vhmmmwmflmammmm.mdfiwwu . 1m .2 . . , . . .1 158.11....: .9 v6 um... I .. . ..nvr JEN. 3%.... . . i - 2. 2a... , .3 .. (.5 PL .5325! ..:}...vlfln. ..:. b21742”. . . .1 1 3:21.53-.. «..-:. . v91... 3. 21.72%... I. ..1. hour»...- "$.15“... 41.) t 4.9‘Uin‘bi 1'... it... . ...l .2 . . . . i - ..Lciui:..9...? ,. .. 3.1.: . E. , 1.. .1! s12..- ...:. .2. U .- -O 11-17.! a‘- "g .. autism... SIN»... .154". ...:... ‘ .4'I a . II I O .3.....E..,..v..- 32...“. \ . 112 5.2 3.1111113Luflh‘fl11 . . .5 )l-u-..xl~luav...-‘51J0 .5- I .HL.‘~I0 no.1vif..|-rf. .. .1! ...ll“: {til-kc 49"tibvuxvv. 1..-, .1 5.1.3020. . ...:... x. 0!. . 6’. UN. .0 ...:-$31 . ..o 11“.... 1.6!. ..LUn no... 1;! . . 1. A3.|-. \.. . . .. . Q I. . .. . f.w.>;.ath!€1.... .11! . x. 61.1.- ..-Xo'lus: and . .126 1 1-1!! 3.3.1.3111 lift. 19‘!!! 4.." . £1 :1- .11-.-. . 1.1-112-.. haw (1.1.129... 31.711111. -. . ..{V ..:. V0.5 hrfihVJ-rirl Viilcal. . _ : .X’I‘ANVI‘. ..u {invitinvv- I" 2 . .1. llfiv \ .b v a 1S3... .. .. “cumin... 1 5 2 5 .‘UI..-. 'II- II- tl.... ...::n.S . . .. al..9>l«:.. 1.x». - '4. O >\ I In . 4333.10. 1411.1. 1).. . . MWMHMDIU. 51.111. .Iuwtule. I‘OQK 20‘: Ill“ . . . . . linur . . . . . . ; : 1 v ..I- - u “AubrhuuV‘. . .. y L . . . . .{ol . -.Il . ’L J . . 1 1‘1".'|l‘ I IVA . .. .1. . . . . . . .. .42 ..I... . . . . ., . H , . .. . “L. . 1 .311..|l. .. .7 . 119.11. lira... ..:. . .. . 6.1.11.1. 0.11.51... .1 \ '11 IL ‘I.! if... .1 19. 12‘ .3311: 41.91"...“rrful... 5...... .. I c. -9 .v. sxl.;n»..1,vl.r. ._ . ...V ntl...‘;$u;a. {.lx ,ltl. I virilk.:l. I.I ‘J‘....Du! 1 . Jul .. ..I... Iltl..l" .I ‘11... . 401 . . IODt'r. 1 v 3 1 THESIS MICHIGAN STATE Li H l! 12” Ill!!!”HHUIIUUNIHI 01579 6596 l LIBRARY Michigan State University This is to certify that the dissertation entitled The Link Between Stereotype Attitudes And Behavioral Intentions Among Female Inmates, Correctional Officers, and Program Staff presented by Pamela Jean Schram has been accepted towards fulfillment of the requirements for Doctor of Philosophy degreein SSC Interdisciplinary Doctoral Program with a Concentration in Criminal JUstice and Criminology Major pr essor Date December 2, 1996 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 PLACE N RETURN BOX to move this checkout from your record. To AVOID FINES return on or bdoro dd. duo. DATE DUE DATE DUE DATE DUE 'JLN Q 9‘ 12378 i. The Link Between Stereotype Attitudes And Behavioral Intentions Among Female Inmates, Correctional Officers, and Program Staff By Pamela Jean Schram A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY School of Criminal Justice 1996 ABSTRACT Within Western cultures, women have been stereotyped into dichotomized images of "good" or "bad." They have also been stereotyped as having limitations or negative traits due to their gender. These cultural stereotypes of women have pervaded the criminal justice system as well. If the criminal justice system is an agent of social control, and stereotypes have permeated the criminal justice system, then to what extent are stereotypes a form of social control over female inmates? This research focused on stereotypes that people have of female inmates and the affects of such attitudes that are pertinent to the social control of women (on performing or not performing certain behaviors). Using survey research methods, four different groups were compared on their attitudes of female inmates. The four groups were: a) female inmates; b) female inmates working in the prison as peer counselors; c) correctional officers; and d) prison program staff. One general attitude toward women was measured in addition to three specific attitudinal measures: a) attitudes about the types of vocational programs for inmates, b) attitudes about motherhood and mothers in prison, and c) paternalistic attitudes toward female inmates. Various measures of behavioral intentions hypothesized to be related to these specific attitudes were also measured. Participants were questioned about how they would respond to various scenarios pertaining to situations of vocational programming, parenting, and paternalism. The results revealed that group membership was significantly related to a general sexist attitude toward women. Additionally, group membership was related to specific attitudes (stereotypes) toward female inmates. When examining the relation between attitudes and behavioral intentions, the results revealed that in some cases the specific attitudinal measures pertaining to female inmates was the best predictor of a respondent's behavioral intentions. In other instances, the best predictor of a respondent's behavioral intentions was her or his general attitude toward women. Copyright by Pamela Jean Schram 1 996 its new rx‘ v3 ‘dr V'- '8'.“ N z: ‘ 1"] ~‘ V. ACKNOWLEDGMENTS There are a number of very special people that have helped me throughout this endeavor. First, I want to thank Dr. Merry Morash. What I have learned from Dr. Morash not only has helped me during my graduate years at Michigan State, but she has shared insights with me that will continue to influence my future endeavors. She has given me guidance but also gave me the opportunity to learn about myself. I also want to thank Dr. Tim Bynum who gave me support and “words of encouragement” when I needed them the most. I want to thank Drs. Janet Bokemeier and Stan Kaplowitz. I greatly appreciate the time and effort they put into being on my dissertation committee. I am extremely thankful to the many graduate students at Michigan State. Barbara Koons was always there with a listening ear and a cup of coffee. Angela Prewitt could make me laugh at times when I really needed it. John Burrow was an extremely patient and understanding office-mate. Other supportive and wonderful graduate students, and friends, include Tracey O’Connell, Victor Bumpus, Steve Cox, Gwen Bramlet Hecker, Kevin Gray, and Donald Hummer. I would also like to thank the many people at the Bedford Hills Correctional Facility. Without their help I would not have completed this project. I would especially like to thank Elaine Lord, Superintendent; Deputy Joseph Smith; Ed Adler, Supervisor of the Counseling Unit; and Pat Johnson, Counselor. Finally, I want to thank my family, especially my Mom and Dad. My parents have always been there for me when I needed them the most. They have always encouraged me to pursue my goals, but they have also helped me to remember the important things in life. It means a great deal to have a wonderful and caring family that is there for me during my fortunate, as well as challenging, moments. ii TABLE OF CONTENTS LIST OF TABLES ........................................ viii LIST OF FIGURES ........................................ xii CHAPTER ONE ........................................ 1 Introduction ........................................ 1 The Current Research ............................. 7 CHAPTER TWO LITERATURE REVIEW ................................... 11 Stereotypes ........................................ 12 What Are Stereotypes? ........................ 12 Study of Stereotypes ........................ 14 Group Membership ............................. 16 Social Distance .............................. 19 Stereotypes and Social Control ........................ 22 Cultural Stereotypes of Women .................. 24 The Connection Between Stereotyping of Women in Prison and Prison Experience .......... 27 Historical Overview of Women in Prison ....................... 28 Today's Female Inmate ....................... 31 Vocational Programs .................. 31 Parenting ............................. 35 ' Paternalism ............................. 4O Attitude-Behavior Relation ....................... 44 Attitude-Behavior Consistency .................. 45 Methodological Approach ....................... 46 Multivariate or "Other" Approach ............................. 47 Pilot Study ...................................... 52 Bedford Hills Correctional Facility ..................... 52 Historical Overview ........................... 53 Bedford Hills Correctional Facility Today ............................ 53 RIV CHAPTER THREE RESEARCH METHODS ................................. Sample ...................................... Survey Instrument ................................. Background Information ....................... Social Distance ........................... Attitudinal Measures ...................... Sexist Attitudes Toward Women Scale ...................... Attitudes about Non- Traditional Work for Female Inmates ....................... Vocational Scale ....................... Parenting Scales ....................... Paternalistic Scales ..................... Behavioral Measures ....................... Scenario One ....................... Scenario Two ....................... Scenario Three ....................... Scenario Four ....................... Scenario Five ....................... Scenario Six ......................... Analysis of the Data ............................ CHAPTER FOUR GENERAL CHARACTERIZATION OF THE FOUR GROUPS OF STUDY ....................... Female Inmates ................................. Children of Female Inmates .................. EducationalNocational Employment Background ...................... Offender History ............................ Peer Counselors ................................. Children of Female Inmates .................... EducationalNocational Employment Background ...................... Offender History .......... Peer Counseling Responsibilities Comparison to National Profile Correctional Officers .......... oooooooooooooooooo 000000000000000000 Employment Background ....................... Program Staff ................................. 105 Employment Background ....................... 107 CHAPTER FIVE GENERAL FACTORS INFLUENCING A'ITITUDINAL MEASURES ........................... 114 Sexist Attitudes Toward Women ...................... 114 Attitudes About Non-Traditional Work for Female Inmates ........................... 1 15 Female lnmates' Need to Work ...................... 1 15 Maintaining Ties With Children ...................... 116 ldealizations of Motherhood ...................... 118 Child Visitation Scale .......................... 118 Paternalistic Attitudes Toward Female Inmates ................................ 1 18 Female Inmates Are Irrational ...................... 120 Summary ...................................... 121 Attitudes About Male and Female Inmates ........................... 124 CHAPTER SIX GROUP MEMBERSHIP AND STEREOTYPES OF FEMALE INMATES ................................ 129 Group Membership and Social Distance — Female Inmates ....................... 129 Sexist Attitudes Towards Women ...................... 130 Attitudes About Non—Traditional Work for Female Inmates ............................ 132 Female lnmates' Need to Work ....................... 134 ldealizations of Motherhood ....................... 137 Maintaining Ties With Children ....................... 138 Child Visitation Scale ............................ 140 Paternalistic Attitudes Toward Female Inmates .................................. 143 Female Inmates Are Irrational ....................... 146 Summary ....................................... 150 CHAPTER SEVEN STEREOTYPES OF FEMALE INMATES BEHAVIORAL INTENTIONS ............................ 153 Scenario One ................................. 153 Scenario Two ................................. 156 Financial Independence ....................... 157 Work Inside the Home ....................... 158 Work Outside the Home ....................... 158 Scenario Three ................................. 159 Scenario Four ................................ 161 Encourage Maintaining Ties .................. 162 Discourage Maintaining Ties .................. 162 Scenario Five ................................ 163 Scenario Six ..................................... 165 Personal Involvement ........................ 166 Paternalistic Treatment ...................... 166 CHAPTER EIGHT DISCUSSION AND CONCLUSIONS ....................... 168 Group Membership and Attitudinal Measures ........................... 168 Attitudinal Measures and Behavioral Intentions ........................... 171 Vocational Programming ....................... 171 Mothers in Prison and Motherhood ................................. 172 Paternalistic Treatment ...................... 173 Theoretical Implications ............................ 174 Policy and Program Implications ....................... 177 Limitations and Recommendations ..................... 180 APPENDICES Appendix A Consent Form ...................... 182 Appendix B Part I - Female Inmates .................. 183 Appendix C Part I - Peer Counselors .................. 186 Appendix D Part I - Correctional Officers ............ 190 Appendix E Part I - Program Staff .................. 192 Appendix F Social Distance Scale .................. 194 Appendix G Sexist Attitudes Towards Women Scale ....................... 195 Appendix H Attitudes about Non-Traditional Work for Female Inmates .................. 198 Appendix I Attitudes about Female Inmates ............................ 199 Appendix J Attitudes about Male Inmates ............................ 203 Appendix K Scenario One ....................... 207 vi Appendix L Scenario Two Appendix M Scenario Three Appendix N Scenario Four Appendix 0 Scenario Five Appendix P Scenario Six . . . LIST OF REFERENCES ........ ooooooooooooooooooooooooo vii 208 209 210 211 212 213 la la Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 11 Table 12 Table 13 LIST OF TABLES Summary of Sample Groups ................. Factor Loadings for "Female Inmates Need to Work Scale" .......................... Factor Loadings for "Maintaining Ties With Children Scale" .................... Factor Loadings for "ldealizations of Motherhood" .......................... Factor Loadings for "Child Visitation Scale" ............................... Factor Loadings for "Paternalistic Attitudes Toward Female Inmates" ............................... Factor Loadings for "Female Inmates Are Irrational" .......................... Summary of Attitudinal Measures and Coefficient Alphas .................... Scenario Two - Financial Independence .......................... Scenario Two - Work Inside the Home .......................... Scenario Two - Work Outside the Home .......................... Scenario Three .......................... Scenario Four - Encourage Maintaining Ties .......................... viii 62 69 70 71 72 73 74 74 76 77 77 78 79 ~Uh T» -C~ I) ) It id ll~ Table 14 Scenario Four - Discourage Maintaining Ties ............................. 79 Table 15 Scenario Five ............................. 80 Table 16 Scenario Six - Personal Involvement ........................ 81 Table 17 Scenario Six - Paternalistic Treatment ........................ 82 Table 18 Summary of Scenario Measures and Coefficient Alphas ........................ 82 Table 19 Summary and Comparisons of Background lnforrnation ........................ 1 12 Table 20 Summary of the Effects of Sex On Attitudinal Measures ........................ 122 Table 21 Summary of the Effects of Age On Attitudinal Measures ........................ 123 Table 22 Summary of the Effects of Educational Level on Attitudinal Measures ..................... 124 Table 23 Comparing Specific Attitudinal Measures Between Female Inmates and Male Inmates ............................ 126 Table 24 Social Distance Among Group Membership ............................ 131 Table 25 Sexist Attitudes Towards Women By Group Membership and Social Distance ........................... 133 Table 26 Attitudes About Non-Traditional Work for Female Inmates by Group Membership and Social Distance ................ 134 la la T. ll Table 27 Table 28 Table 29 Table 30 Table 31 Table 32 Table 33 Table 34 Table 35 Table 36 Table 37 Table 37 Mean Group Scores for Individual Test Items on the Female lnmates’ Need to Work ................................. Female Inmates’ Need to Work by Group Membership and Social Distance ............................ Mean Group Scores for Individual Test Items on ldealizations of Motherhood .................................. ldealizations of Motherhood by Group Membership and Social Distance Mean Group Scores for Individual Test Items on Maintaining Ties With Children ...................... Maintaining Ties With Children by Group Membership and Social Distance ...... Mean Group Scores for Individual Test Items on Child Visitation Scale . . . . Child Visitation Scale by Group Membership and Social Distance ...... Mean Group Scores for Individual Test Items on Paternalistic Attitudes Toward Female Inmates ...... Paternalistic Attitudes Toward Female Inmates by Group Membership and Social Distance ...... Mean Group Scores for Individual Test Items on Female Inmates Are Irrational ...................... Female Inmates Are Irrational by Group Membership and Social Distance ...... nnnnnnnnnn 135 136 138 139 141 142 143 144 145 147 148 148 Tea? .1. AF ..ilill lac; Table 39 Table 40 Table 41 Table 42 Table 43 Table 44 Table 45 Table 46 Summary of the Effects of Group Membership and Social Distance on Specific Attitudinal Measures ................ Summary of Regression Analyses ........... Scenario One: Encourage Participation In Non-Traditional Vocational Programming ........................... Scenario Two: Financial Situation ........... Scenario Three: Encourage Mothers In Prison to Retain Custody ................ Scenario Four: Maintaining Ties With Children ........................... Scenario Five: Involvement in an Emotional Situation ..................... Scenario Six: Paternalistic Situation ........... xi 151 154 155 159 161 163 164 167 Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 LIST OF FIGURES Conceptual Framework ........................ Age Distribution of Female Inmates ............. Racial/Ethnic Background of Female Inmates ............. Age Distribution of Children ..... Children Living At Home Prior To Mothers' Incarceration (18 Years or Younger) ....................... Educational Background ....... Types of Vocational Programming Primary Source of Income ....... Current Offense ............. Minimum Sentence Serving . . . Age Distribution of Peer Counselors Racial/Ethnic Background of Peer Counselors ............ Age Distribution of Children .................. Children Living At Home Prior To Mothers Incarceration (18 Years or Younger) ........................ Educational Background ........................ xii 23 88 88 89 89 91 91 93 93 95 97 97 99 Figure 16 Figure 17 Figure 18 Figure 19 Figure 20 Figure 21 Figure 22 Figure 23 Figure 24 Figure 25 Figure 26 Figure 27 Figure 28 Figure 29 Figure 30 Figure 31 Figure 32 Figure 33 Types of Vocational Programming Primary Source of Income Current Offense Minimum Sentence Serving Peer Counseling Responsibilities Age Distribution of Correctional Officers Racial/Ethnic Background of Correctional Officers Educational Background Correctional Officers Work Shift Age Distribution of Program Staff Racial/Ethnic Background of Program Staff Educational Background Employed at a Male Facility Services Provided to Inmates Sex and Attitudes About Female Inmates' Need to Work ...... Sex and Attitudes About Maintaining Ties With Children Sex and Attitudes About ldealizations of Motherhood Sex and Attitudes About Child Visitation xiii ooooooooooooooooooo 00000000000000000 100 100 102 102 104 104 106 106 108 108 110 110 111 117 117 119 119 Fe Figure 34 Sex and Attitudes that Female Inmates Are Irrational ....................... 120 Figure 35 Mean Social Distance Scores ................. 131 Figure 36 Mean Sexist Attitudes Toward Women Scores ............................ 133 Figure 37 Mean Female Inmates' Need To Work Scores ............................ 136 Figure 38 Mean ldealizations of Motherhood Scores ....................... 139 Figure 39 Mean Maintaining Ties With Children Scores ....................... 142 Figure 40 Mean Child Visitation Scores ................. 144 Figure 41 Mean Paternalistic Attitudes Toward Female Inmates Scores ................. 147 Figure 42 Mean Female Inmates Are Irrational Scores ............................ 149 xiv CHAPTER ONE INTRODUCTION In 1922, Lippman introduced the concept of stereotypes as "pictures in our heads." Ashmore and Del Boca (1981) defined stereotypes as "[a] set of beliefs about personal attributes of a group of people" (p. 16). Although there is general agreement regarding this "core meaning," there is disagreement about the more specific characteristics of stereotypes. Stereotypes have been studied from three basic theoretical orientations: 1) the psychodynamic; 2) the cognitive; and 3) the sociocultural (Ashmore & Del Boca, 1981; Hamilton, 1979). The psychodynamic and cognitive orientations focus on personality and processes of stereotyping, respectively. The sociocultural orientation emphasizes the content of stereotyping (Ashmore & Del Boca, 1981; Hamilton, 1979, 1981; Rothbart, Fulero, Jensen, Howard, & Burrell, 1978). Within the sociocultural orientation, "stereotypic beliefs [are] a reflection of the cultural or social milieu which has shaped one's experience and Ieaming history" (Hamilton, 1979, p. 55). The sociocultural orientation guides the present study because of the focus on societal influences of an individual's stereotypes rather than on an individual's psychological processes of stereotyping. Thus, instead of examining psychological explanations of stereotyping, this research examines sociological 2 explanations. Research emanating from this orientation is frequently a study demonstrating “that a set of perceivers agree about the characteristics of some target group or groups" (Ashmore & Del Boca, 1981, p. 24). Essentially, this type of research has focused on gender1 and ethnic stereotypes (e.g., Ashmore & Del Boca, 1979; Biemat & Crandall, 1994; Branscombe & Smith, 1990; Fairchild, 1985; Huddy, 1994; Jackson, Sullivan, & Hodge, 1993; Jones, 1991; Lalonde 8. Gardner, 1989; Morrison, Bell, Morrison, Murray, & O'Connor, 1994; Pratto 8 Bargh, 1991; Rickman, 1983, Sigelman, Sigelman, Walkosz, & Nitz, 1995; and St. Pierre, Herendeen, Moore, & Nagel, 1994). Research on gender stereotypes emphasizes primarily, although implicitly, the sociocultural perspective: "Sex [gender] stereotypes are assumed to be part of the same cultural pattern that specifies sex [gender] roles and sex-[gender-1role standards" (Ashmore & Del Boca, 1981, p. 23). Within this orientation are two types of models: the structuralist- functionalist and the conflict. The structuralist-functionalist model is based on the premise that cultural stereotypes are essentially maintained due to consensual agreement. However, the conflict model, which is particularly helpful in understanding gender stereotyping in prison, conceptualizes stereotypes as a method for justifying or rationalizing existing patterns of intergroup relationships. This model can provide the necessary framework for understanding how various groups within the prison setting can maintain opposing and conflicting is: fan I... r . 5) ‘5‘? 3 stereotypes of female inmates. Jackson and Senter (1980) contend that attributing certain characteristics to group membership can "provide the cognitive foundation for social policies and practices that affect the life changes of individual group members and the patterns of relationships between groups" (p. 341 ). Women have experienced being assigned characteristics due to their "group membership." From antiquity to the present, cultures have categorized women into "either-or“ roles (Pomeroy, 1975, p. 8). One such pervasive conceptualization was the madonnalwhore duality. Another similar and pervasive conceptualization was femininity (Dugger, 1991 ). Several writers contend that this cultural ideology has been reflected and perpetuated by the criminal justice system (Belknap, 1996; Feinman, 1980; Klein, 1982; Rafter, 1 990). Stereotypes of female offenders are not created within a vacuum. Rather, these stereotypes are reflected and perpetuated by society's assumptions as to whom will be deemed a "true" or a deviant woman. Price and Sokoloff (1982) argued that the criminal justice system consists of various institutions which are components of a larger society. As such, "the system both reflects and perpetuates society's values" (p. xii). If society perpetuates sexist assumptions, such biases can be found in the criminal justice system as well. The criminal justice system consists of numerous agents, at various stages of processing, with the authority to establish as well as enforce the rules is: “5. Ad an: 4 of society. The criminal justice system is a system of social control or one approach "in which society responds to behaviour and people it regards as deviant, problematic, worrying, threatening, troublesome or undesirable in some way or another" (Cohen, 1985, p. 1). A theoretical framework of social control provides the necessary perspective to explain these processes. The literature has proposed various conceptualizations of the construct "social control." In most instances, these conceptualizations were identified by designating the means in which social control is implemented (e.g., Amir 8. Biniamin, 1991; Cohen, 1983; Etzioni, 1961; Gagne, 1992; Green, Hebron, 8. Woodwards, 1987; Lengerrnan 8. Wallace, 1985; Smart & Smart, 1978). Smart and Smart (1978) argued that the social control of women exists in various forms such as internal or external; implicit or explicit; and private or public. Smart and Smart further contend that [t]he social control of women assumes many forms, it may be internal or external, implicit or explicit, private or public, ideological or repressive. Now although it may no longer be appropriate to talk of 'the problem that has no name' when referring to the discontents of women, the Women's Movement having provided a voice and a language with which women may articulate their manifest grievances, there remains the problem of showing the existence of specific covert forms of oppression and control, and of revealing that their location lies in the public sphere rather than in the individual psychologies or personal lives of oppressed women [italics in original] (p. 2.). In her study of Appalachian women, Gagne (1992) identified one form of social control as "normative control." This type of social control consists of nonactive and nonviolent forms. It is a result of cultural norms, values and beliefs - and I85 (A) (( iii (2 £? 5 related stereotypes - as well as the social structure. Normative control is the result of both the patriarchal social structure as well as biased cultural beliefs (p. 392). There have been criticisms of social control theories (Chunn & Gavigan, 1988; Cohen, 1983). Cohen (1983) argued that the constmct of "social control" has become a "Micky Mouse" concept implemented to explain a wide range of social processes. Chunn and Gavigan (1988) seriously questioned the conceptual and analytical adequacy of the construct of social control. They concluded that the construct has a limited utility "for developing an historically and theoretically informed understanding of the complex and the contradictory relationship of women to the state and IaW' (p. 120). Rather than abandoning the concept of social control, one needs to re-think, re-conceptualize, and re- evaluate the concept. An initial step to re-evaluating the concept of social control is to ask, 'What are the extent and forms of social control that women experience in the criminal justice system?" One may further ask, "If stereotypes can be found in the criminal justice system, and, if the criminal justice system is a mechanism of social control, are stereotypes a form of social control over female offenders?" Several writers have argued that historically, women in prison have been negatively affected by stereotypes (Dobash, Dobash, & Gutteridge, 1986; Feinman, 1983; Fox, 1984; Mann, 1984; Pollock-Byme, 1990; Rafter, 1990; Zupan, 1992). Historically, most female offenders were considered to have loos fare 8816 etc 6 violated society's moral standards, especially standards related to sexuality, rather than standards of law-abiding behavior. Women involved in criminal activity were worse than men because they not only sinned, but they also loosened the moral constraints on men. Because women were born pure, female offenders were more depraved than male offenders. This justified the severe treatment of female criminals (Freedman, 1981 ). In this vein, Feinman (1983) asserted that during the urbanization and industrialization period of the United States, the Cult of True Womanhood extolled the virtues of femininity (Carlen, 1982; Smith, 1990; Welter, 1973). These cultural stereotypes of women, "true womanhood," influenced the treatment of female offenders. An essential component of this movement was that female offenders should be "treated" differently than male offenders. These philosophies and assumptions greatly influenced the form of treatment women received in prison. The reformers perceived the goal of rehabilitation as "molding" these women to be good homemakers and mothers. Are these historical trends in any way reflected in the contemporary treatment of female prisoners? Some criminologists assert that the lack of understanding and negative stereotypes regarding women offenders is still prevalent today and "continue to intrude on the treatment of women in prison" (Feinman, 1979, p. 132). The literature has provided various examples of how such stereotyping can affect attitudes concerning a) the types of vocational programs offered to inmates; b) motherhood and mothers in prison; and c) 7 paternalistic attitudes toward female inmates. Some have argued that vocational programs for female inmates continue to focus primarily on reinforcing a traditional role perspective (Carlen, 1982; Carp & Schade, 1993; Chapman, 1980; Morash, Haarr, & Rucker, 1994; Moyer, 1984; Simon & Landis, 1991; Weisheit, 1984). ldealizations of motherhood can influence attitudes concerning mothers in prison (Beckennan, 1994; Mahan, 1982; Mann, 1984). Others may not realize that upon release many mothers are the primary caretaker (Datesman 8. Cales, 1983; Henriques, 1982). Some writers have insisted that perceiving female inmates as though they were children and highly irrational results in paternalistic treatment (Burkhardt, 1973; Fox, 1984; Freedman, 1981; Pollock, 1984; Pollock-Byme, 1991). THE CURRENT RESEARCH Criminological literature has made claims and provided descriptive evidence that such stereotyping does exist and negatively impacts women in prison (Feinman, 1979; Hutter & Williams, 1980; Pollock-Byrne, 1991; Smart, 1977). However, there is a paucity of research that precisely examines who stereotypes whom as well as how these stereotypes are linked to behavior towards women, and the more general phenomena of social control. A lack of understanding and examination makes it difficult to assess how stereotypes shape correctional programming and practices. Thus, it is difficult to pose any solutions or correctives. Furthermore, there is a lack of understanding about. Inmate disilnc term 9130.91 Delia inmate :32be 8 about whether various groups (i.e., correctional officers, program staff, and inmates themselves) have differing attitudes and behaviors possibly due to their distinct function within the institution. This study attempts to provide a more in- depth understanding of stereotypes and behavioral intentions among various groups within a women's correctional facility, and examines the relationship between stereotypes and the intended behavioral interactions with female inmates. In the theoretical framework for the present research, stereotypes are a form of social control. If stereotypes promote and limit women to a narrow range of behaviors and roles, they exert social control over them. If powerful or influential agents in the prison system hold such attitudes, then the resulting expected behaviors are a form of social control for the female inmates. However, one may further question the extent that the female inmates themselves also hold stereotypes towards other female inmates. To examine if such stereotypes exist, a survey method was implemented (Babble, 1990; Rea & Parker, 1992). Female inmates, as well as correctional officers and program staff, were surveyed. First, a social distance scale was administered to assess if group membership influenced the type of interaction between these various groups. Second, attitudes of these groups concerning vocational programming, parenting, and paternalism were examined. Finally, to assess if such attitudes were related to intended behavioral interactions with female inmates, respondents were provided with six scenarios pertaining to 9 vocational programming, parenting, and paternalism. Participants were asked to provide hypothetical behavioral responses to these scenarios. ENDNOTES 1This research recognized the distinction between the terms "sex" and "gender" stereotyping. Sex is determined by "socially agreed upon biological criteria for classifying persons as females or males (West & Zimmerman, 1991, p. 14). This criteria is based on physiological differences between women and men. Gender is determined by socially agreed upon activities and attitudes about individuals according to their sex This study examined gender stereotypes. Specifically, this study focused on attitudes concerning various activities, as well as images, emanating from the social construction of gender. However, while this research recognized the distinction between sex and gender, some of the studies cited did not make such a clear distinction. 10 TB‘E CHAPTER TWO LITERATURE REVIEW The following review integrates literature that explores various factors relevant to stereotypes of female inmates. The first section provides a general discussion of stereotypes and then focuses on cultural stereotypes of women which often categorize women into dichotomous groups (i.e., "good" or "bad"). After a brief and general discussion of the social control perspective, the next section provides an argument that stereotypes of female inmates is a form of social control. Such stereotypes can influence the types of vocational programs offered to inmates, motherhood and mothers in prison, and paternalistic attitudes toward female prisoners. If such stereotyping exists, to what extent do these attitudes influence the behaviors of female inmates, correctional officers, and program staff? The next portion of this discussion addresses the general issue of attitude-behavior consistency and the various approaches that have been suggested to increase the connection in research. The specific approaches utilized for this study are also discussed. The final portion is a brief overview of the female correctional institution that was selected for this study. 11 1 2 STEREOTYPES As noted in the introductory chapter, the term "stereotype" was introduced to social science in 1922 in Lippman's Bummer]. He conceptualized stereotypes as "pictures in our heads": [Pjictures inside the heads of these human beings, the pictures of themselves, of others, of their needs, purposes, and relationships (p. 29). Ashmore and Del Boca (1981) provided the following "core meaning" of stereotype: A set of beliefs about the personal attributes of a group of people. It is important to note that we are not proposing a single best conceptual definition of the stereotype construct; rather, we are attempting to state explicitly the essential defining features of the term as it has been used by social scientists [italics in original] (p. 16). Many generally agree with this definition of stereotypes. However, there is less agreement concerning specific conceptual issues relevant to stereotypes. For instance, a) does the construct stereotype itself have an "inherent badness" within the definition? and b) are stereotypes more of an individual belief, or a cultural belief, about certain social groups? (Ashmore & Del Boca, 1981, p. 21) What Are Stereotypes? Although stereotypes may be characterized as "bad," this "badness" should not be incorporated in the concept of stereotypes. Three reasons were offered for maintaining this distinction: a) parsimony in defining scientific terms 1 3 should not be the only optimal value in this instance because "there is little to be gained and much to be lost by making badness a point of definition (p. 16)"; b) when a judgement, such as "bad," is associated with the definition of "stereotypes," this infers that stereotypes and stereotyping are deviant or pathological cognitive structures and processes; and c) when such inferences are made, research on stereotypes is disassociated from relevant "basic" research and theory that could attempt to study the alleged reasons for badness rather than assume its existence. In reference to the second conceptual issue, Ashmore and Del Boca (1979) argued that this contradiction in opinion centers on the question of consensus. They distinguished the concept of "stereotype" and "cultural stereotype." Stereotype was designated as individual beliefs concerning a social group while "cultural stereotype" was delineated as a consensual set of beliefs on a societal level. Distinguishing these two constructs was emphasized especially in reference to research questions. For instance, "Are there differences in the degree to which individuals adopt cultural stereotypes?" (p. 19) The focus of this research recognizes that the construct of stereotype, generally, is not associated with any negative value judgments, or "badness." While the discussion below demonstrates that some prior writers have concluded that there are negative stereotypes associated with female inmates, the present research does not assume, decisively, that such negative 14 stereotypes do exist. Furthermore, this research realizes the distinctions between "stereotype" and "cultural stereotype." This issue of consensus is an essential interest in this research. By adapting Ashmore and Del Boca's (1981) question mentioned above, this research focuses on assessing the degree to which female inmates, correctional personnel, and program staff adopt cultural stereotypes of female inmates. Study of Stereotypes As mentioned previously in the introductory chapter, the study of stereotypes has essentially emanated from three basic orientations: 1) the psychodynamic; 2) the cognitive; and 3) the sociocultural (Ashmore & Del Boca, 1981; Hamilton, 1979). However, "[these orientations] are not 'theories', rather, they are frames of reference that guide, often quite implicitly, the conduct of research" (Ashmore & Del Boca, 1981, p. 22). The psychodynamic orientation emphasizes how stereotypes are associated with prejudice and personality. The cognitive perspective focuses on the processes of stereotypes. The sociocultural perspective emphasizes the content of stereotyping (Ashmore & Del Boca, 1981; Hamilton, 1979; 1981; Rothbart, Fulero, Jensen, Howard, & Burrell, 1978). As stated above, these orientations guide ”the conduct of research." The sociocultural perspective is the frame of reference implemented for the proposed 1 5 research. Within the sociocultural orientation, the primary focus of interest is society: Individuals are socialized into a particular culture, and, through social rewards and punishments, led to act in accordance with cultural dictates. Further, by accepting cultural stereotypes, individuals reinforce and thereby help to perpetuate the existing cultural pattern (Ashmore & Del Boca, 1981, p. 23). Ashmore and Del Boca (1981) identified two types of models emanating from this orientation: the structuralist-functionalist and the conflict. The former is founded on a societal consensus view while the latter is based on differing, competitive values and interests. These writers further explained that the conflict perspective maintains stereotypes are a method for justifying or rationalizing existing patterns of intergroup relationships. Cultural stereotypes are implemented as a value- expressive function. An example of mentally disabled individuals was provided to illustrate this function. Negative stereotypes of such individuals reinforce cultural values regarding intelligence and independence. Individually, expressing stereotypes can confirm societal values which a person may identify to be similar to his or her own. Thus, by expressing these beliefs, an individual attains social acceptance (p. 24). Ashmore and Del Boca (1981) noted that "[t]he most frequent type of study is a simple demonstration that a set of perceivers agree about the characteristics of some target group or groups" (p. 24). As noted in the introductory chapter, this pattern of research has essentially focused on gender and ethnic stereotypes. 16 Group Membership Jackman and Senter (1980) argued that the assignment of certain traits to group membership can "provide the cognitive foundation for social policies and practices that affect the life changes of individual group members and the pattern of relations between groups" (p. 341). In their article, Jackman and Muha (1984) examined how educational attainment influenced intergroup attitudes. One alternative interpretation of their study was to incorporate the concept of dominant ideology: Students of social inequality and group consciousness have long regarded ideology as integral to relations of inequality. Groups that occupy a dominant position in the social structure routinely manufacture an interpretation of reality and a set of normative prescriptions that serve their interests. . . . Dominant groups develop such an ideology without contrivance: it flows naturally from their side of experience as they seek to impose a sense of order on the pattern of social relations and to persuade both themselves and their subordinates that the current organization of relationships is appropriate and equitable (p. 759). According to their argument, stereotypes of intergroups are perpetuated by a "power" motivation - an attempt to maintain the privilege of the dominant group over the subordinate groups. Jackman and Muha (1984) further reasoned that intergroup attitudes of dominant groups are not uncontrolled outbursts of negativism; nor are they the anachronistic expressions of deficiencies in socialization or personality. Instead, they are an important part of dominant groups' attempts to control the social relations from which they benefit. Dominant groups thus refrain from unduly offending the sensibilities of themselves or subordinates (p. 759). 1 7 When "harmonious inequality" is questioned by subordinates, those comprising the dominant groups attempt to disquiet such challenges to the status quo by implementing a subtle, but effective, form of control. This form of control is to implement a persuasive, adversarial posture that will pacify rather than incense those in the subordinate groups (p. 759). In reference to the present study, while some individuals may have negative stereotypes of female inmates, such stereotypes may not be deemed as "uncontrolled outbursts of negativism." Rather, they are designated as negative because of the potential of limiting women to certain behaviors and roles without providing a broader range of options. Jackman and Muha (1984) concluded that an essential and purposive component of dominant ideology is to maintain the interests and benefits of the dominant group. Simultaneously, this safeguarding of dominant interests is maintained by avoiding or circumventing challenges raised by the subordinate groups. However, if challenges are raised, the alternative approach is to lessen group characteristics: Group distinctions are minimized, and dominant groups develop a commitment to individual rights both as a diversion from subordinate-group demands and as a principled basis for the rejection of such demands. In the face of subordinate challenge, dominant groups also become practiced in the art of making symbolic concessions in lieu of tangible concessions (p. 765). Jackman and Muha argued that the educational system was examined as a major vehicle for attempting to distort the protection of dominant ideology. In this 18 vein, a possible "symbolic concession" to arguments that female inmates do not receive an adequate number of vocational programs, would be to implement additional programs that are characteristically gender-biased (e.g., cosmetology). An essential component of dominant ideology is the designation of in- groups and out-groups. Specifically, how are individuals delineated into certain groups? Wilder (1981) focused on how individuals categorize themselves as in- group and, alternatively, how they perceive the out-groups. Of particular interest to this research, Wilder argued the importance of the structural contact between groups and categorization processes. He asserted that when structural boundaries are intact, the recognition of individual characteristics, rather than group membership, is less effective: One would expect contact that allows the group boundaries to remain highly salient (thereby emphasizing the ingrouploutgroup categorization) to be less effective than contact that mixes individuals of both groups. Consider, for example, contact between two groups where each group sits at a separate table wearing identification tags. Although the setting may be designed to induce cooperation between the groups, ingroup identification should be salient throughout the interaction. Such circumstances would encourage a perception of the outgroup members as a single unit, and not as a more individuated set of persons (p. 252). However, by lessening the structural contact among the groups (i.e., the boundaries among the groups are obscured), people may perceive others as individuals rather than group members. Within the prison setting, group membership is highly salient. The need for an individual to identify her or his group membership is essential for the operation of the institution: Who is an inmate? Who is a correctional officer? Who is a program staff member? This group membership identification specifies the appropriate policies and procedures that are to be followed by a group member. If group membership is essential within a prison setting, than there is the potential to establish as well as maintain restrictions on various interactions between these differing groups. These varying interactions between identifiable group members has been referred to in the literature as social distance. Social Distance In no case is the "group" to be thought of as an abstraction, or as something apart from its members. Neither is it a mere conglomeration of individuals. It often acts and feels as a unit, and as such its unified actions in relation to other groups constitute "distance" indexes. Taken in connection with a lengthening time period, the "distance" meaning of these indexes deepens. (Bogardus, 1926, p. 477) As cited in Owen, Eisner, and McFauI (1981), in 1924 Park defined social distance as that distance which applies to human feelings rather than spatial distance: He conceptualized social distance as a mechanism for reducing to measurable terms the amount of understanding and intimacy which characterizes personal and social relations. He noted that it is common practice for people to state feelings of closeness or distance to other people. Park expressed hope that eventually a social distance instrument could be devised which might easily and accurately measure personal feelings toward others . . . (p. 81) 19 20 Bogardus (1925, 1933) developed such a scale to measure social distance which consisted of the following seven items: Would marry Would have as regular friends Would work beside in an office Would have several families in my neighborhood Would have merely as speaking acquaintances Would live outside my neighborhood Would live outside my country (Bogardus, 1933, p. 269) NQS’PP’Nr‘ The scale was developed to measure social distance between various ethnic and social groups. The numerous studies that have administered this scale focused on the relation between educational level and racial/ethnic interactions (Pass, 1988); the comparison of social distance between college students and inmates (Pass, 1987); the relation of social distance and six ethnic groups in Canada (Netting, 1991); the comparison of ethnic groups in Australia (McAIlister & Moore, 1991); the racial, ethnic and social distance in a Caribbean community (Brinkerhoff & Jacob, 1994); racial and ethnic social distance among college students (Sparrow & Chretien, 1993); and the examination of social distance of racial/ethnic groups among "white" sorority and fraternity members (Muir, 1991). As illustrated by the research above, an essential component of social distance is group membership. The present research focuses on four groups: a) female inmates; b) peer counselors; c) correctional officers; and d) program staff. Each group has distinct roles and responsibilities within the institution. 21 The main responsibilities of correctional officers is the maintenance of security and custody within the institution. The responsibilities of the program staff, however, are not to maintain custody within the institution. Rather, they provide numerous social services for the female inmates. By virtue of their roles, their interactions with the female inmates are characteristically different than correctional officers. The program staff and correctional officers were the only groups within this study that were not inmates. The interactions between the program staff with peer counselors may be different than with the female inmates because the peer counselors provide similar duties as the program staff themselves. The responsibilities of the female inmates are essentially dictated to them by the other groups within the prison system. Their interactions with the correctional officers are primarily in connection with maintaining security and custody while their interactions with program staff are to obtain various services. Among this group of female inmates is the group of peer counselors. Peer counselors have a unique niche within the prison system. They are primarily inmates who have received long-term sentences. However, they have an additional responsibility of providing services to the other inmates in the prison. Peer counselors are required to have at least a high school diploma or GED. as well as receive training and certification. As illustrated above, each of these groups have distinct roles and responsibilities within the institution. This differentiation could result in social 22 distance between these groups. However, social distance, in this context, is not in reference to an ethnic or racial group. Rather, it is within an organizational context. For the present study, social distance is conceptualized as the types of interactions that occur within the institutional setting. Thus, group membership was hypothesized to be related to social distance which was conceptualized as interactions (see Figure 1). STEREOTYPES AND SOCIAL CONTROL In their critical work on the social control concept, Chunn and Gavigan (1988) argued that there are two definitions of social control. One definition was designated as benign social control which is premised on the assumption that societal integration is achieved through the operation of numerous non-coercive social control processes. (P- 108) Social control processes include teaching individuals "right" and "wrong" as determined by rules, norms, and values (Williams & McShane, 1988, p. 108). However, these rules, norms, and values are assumed to be accepted rather than challenged. The other definition was designated as coercive social control which is premised on the belief that coercive state control mechanisms, particularly law, play the most crucial role in reproducing the status quo (p. 108). GROUP MEMBERSHIP a) fetnaleinmates b) peer counselors- c) correctional officers d) program staff 9K4; 23 Figure 1 Conceptual Framework Social Distance Behavioral Intentions Y Attitudes ATTITUDES TOWARDS WOMEN a) Encouragement or discouragement to participate in non-traditional programs b) Encouragement or discouragement to be financially independent VOCATIONAL TRAINING a) Type of occupation b) Need to work PARENTING a) 'Bad" mothers b) Separation from children a) Encouragement or discouragement for mothers to retain custody b) Encouragement or discouragement to maintain ties with children PATERNALISM a) Irrational b) Child-like a) Interactions with inmates in a stressful situation b) Treattnent of inmates as children With Men pioce mm 08m Amen. 24 Within this perspective, the rules, norms, and values are challenged, particularly when the "state control mechanisms" have the power to enforce these processes. Cultural stereotypes of women have been embedded in these rules, norms, and values. Cultural Stereotypes of Women From antiquity to the present, cultures have categorized women into "either-or" roles (Pomeroy, 1975, p. 8). One such pervasive conceptualization is the madonnalwhore duality. This duality is grounded on two contrasting perceptions of the female "nature" or sexuality. The madonna image personifies women as faithful and submissive wives as well as nurturing mothers. The whore image portrays a woman as a temptress of a man's sexuality and self- control (Feinman, 1980; Rafter, 1990). The specific characteristics ascribed to women's nature and those critical to theories of female criminality are uniformly sexual in their nature. Sexuality is seen as the root of female behavior and the problem of crime. Women are defined as sexual beings, as sexual capital in many cases, physiologically, psychologically and socially [italics in original] (Klein, 1982, p. 37). This dichotomy, however, has been described from a white, upper-middle class woman's perspective. Young (1986) argued that the categorizations for African American women are not dichotmized as "good" or "bad." Rather, the African American woman "has been characterized as an Amazon, a 'sinister Sapphire,’ a mammy, and a seductress. Unlike characterizations of her white female counterpart, which are either good or bad, all the categorizations of the black 25 female are bad" (p. 322). An Amazon is depicted as domineering, assertive, and masculine, whereas a "sinister Sapphire" is deemed as dangerous and cestrating. A mammy is characterized as the long-suffering paragon of patience, while the seductress is immoral and sexually depraved (p. 323). Another similar and pervasive conceptualization of femininity is that it consists of various traits such as gentleness, sensitivity, nurturance and passivity. Comparatively, traits associated with masculinity include intelligence, aggressiveness, independence and competitiveness (Dugger, 1991). These conceptualizations become problematic when such traits are assumed to be inherent to an individual's sex or are considered as "biological fact“ (Brownmiller, 1984; Edwards, 1989). Borrowing Goffman's terminology, West and Zimmerman (1991) defined gender roles as behavior and role enactment within a social situation. Gender roles are not based on biological differences between males and females. Rather, these roles are socially constructed differences between men and women. As discussed previously, categorization is an essential factor of stereotyping. Rather than stereotyping a person, one stereotypes a person-as- a-group—member (Taylor, 1981). There are two possible ways which a female inmate could be categorized: a) as a woman; or b) as a woman prisoner. To explore attitudes of various groups concerning women generally, the following research question was addressed: 26 Research Question: Is group membership related to attitudes toward wromen? In Figure 1 this research question is depicted by the line connecting Group Membership and the box labeled "Attitudes toward Women," with Social Distance as a controlling factor. Specifically, this question examines if group membership and social distance individually have a significant effect on attitudes toward women. The literature suggests that sex and age significantly influence attitudes toward women. Males tend to have more negative or traditional attitudes toward women compared to females (Harry, 1995; Innes, 1993; Jones & Jacklin, 1988; Street, Kimmel, & Kromrey, 1995; Szymanski, Devlin, Chrisler, 8. Vyse, 1993). Some studies have reported older adults having more traditional attitudes toward women compared to younger adults (Benson 8. Vincent, 1980; Houser & Beckman, 1980; Schroeder, Blood, & Maluso, 1992; Spence & Helmreich, 1979). However, studies comparing younger and older college students have reported age being positively correlated with liberal attitudes toward women (e.g., Etaugh & Spiller, 1989; McKinney, 1987). Since there appears to be some relation between sex, age, and attitudes toward women, these factors have been included in the present study (see Figure 1). 27 The Connection Between Stereotyping of Women in Prison and The Prison Experience In the introductory chapter, an initial question to re-conceptualizing social control had been to ask, 'What are the extent and forms of social control that wemen experience in the criminal justice system?" Heidensohn (1985) argued that there are various forms of social control. These forms of social control are both informal and institutional, and they define as well as limit the behavior of women: Of all the subtler constraints on the way women act and are supposed to act, few are more complex than the workings of social policies. Social policies are not usually regarded as instruments whose prime purpose is the definition and enforcement of prescriptions about gender roles, especially women, but a growing body of analyses shows that such prescriptions underpin, or are an effective part of certain policies (p. 191 ). Smart and Smart (1978) asserted that the more public or "visible" forms of social control have been addressed. However, "the more difficult forms of social control to address, especially with the eclipse of more manifest forms of sexual discrimination are those that arise implicitly through socialization" (p. 2). Continuing with the above question, one may further ask, "If stereotypes can be found in the criminal justice system, and, if the criminal justice system is a mechanism of social control, are stereotypes a form of social control over female offenders?" The following portion of this discussion attempts to answer this question. Initially, the discussion focuses on the historical experiences of 28 women in prison. Thereafter, the discussion explores three contexts in which women, today, continue to experience social control through stereotypes: a) vocational programming, b) motherhood; and c) paternalism. WWW Many correctional practices were attempts to rehabilitate the female offender into an idealized concept of what it means to be a "the woman." This need to rehabilitate or change a woman offender was thought to be due to her lack of proper internalization of society's definition of a "true woman." If a women did not "internalize" these qualities, then she was deviant and needed to change or be changed. Several writers contend that historically, women in prison have been affected by stereotypes regarding femininity (Dobash, Dobash & Gutteridge, 1986; Feinman, 1983; Fox, 1984; Mann, 1984; Pollock-Byrne, 1990; Rafter, 1990; Zupan, 1992). Most female offenders were considered to have violated society's moral standards, especially sexuality, rather than standards of law- abiding behavior. Feinman (1983) asserted that during the urbanization and industrialization period of the United States, the Cult of True Womanhood extolled the virtues of femininity. The attributes of True Womanhood consisted of four virtues: piety, purity, submissiveness and domesticity. (Carlen, 1982; Smith, 1990; Welter, 1973). A woman deemed as "bad" usually had at least one of the following characteristics: a) she was indecisive and lacked "moral 29 fortitude," b) she was promiscuous; or c) she was irresponsible because not only was she loosening her morals and values but those of her mate and descendants as well (Hahn, 1980, p. 3). Women involved in criminal activity were worse than men because they not only sinned, but they also loosened the moral constraints on men. Because women were born pure, female offenders were more depraved than male offenders. This justified the severe treatment of female criminals (Freedman, 1981, p. 78). However, in the nineteenth century, prison reformers began to perceive female offenders as being "misguided" rather than evil, "fallen women" (Pollock-Byme, 1990; Rafter, 1990): Late nineteenth-century beliefs about the nature of women in general helped demote the female criminal from the status of a mature, if wicked woman to that of an impressionable girl. As social class distinctions hardened within nineteenth-century society, middle-class women became "ladies," delicate and vulnerable creatures. No one expected factory girls or domestic servants to display all the attributes of the lady, but in discussions of "women's nature," traits associated with the lady were generalized to all women. Even the female offender was now depicted as frail and helpless, more a vulnerable child than a hard- hearted enchantress (Rafter, 1990, p. 49-50). The history of women in prison illustrates how such philosophies regarding "true womanhood" influenced the treatment of female offenders. For instance, during the mid-18003, the matron of the women's section at Sing Sing, Georgiana Bruce Kerby, wrote about this reform philosophy: As I said, it had pleased us to love these low—down children of circumstances less fortunate than our own. We gloried in being able to lift a few of them out of the slough into which they had 30 fallen, or in which they have been born, and to sustain them while they were trying to take a little step upward in the direction of the light (Feinman, 1980, p. 45). In 1873, the first all-female institution was established, the Indiana Refon'natory Institution for Women and Girls (Rafter, 1990, pp. 29-30). The reformers advocated three goals that would result from the establishment of separate institutions for female inmates: First, sexual abuse and exploitation of female prisoners would be prevented. Second, the female staff would set a moral example of "true womanhood" for the female offenders to emulate. Finally, these staff would provide sympathetic counseling to their charges ’ (Zupan, 1992, 297-298). The primary ideological motivation for separate institutions, as well as separate treatment, was a policy heavily influenced by gender stereotypes: While the reformers obtained separate prisons for women under women administrators and staff, and improved treatment for incarcerated women, they reinforced and perpetuated stereotypical sex roles for women. Women's corrections would be matriarchy where "good" staff women, acting as mothers, would teach the inmate-children to be proper women in a simulated homelike environment in the prison (Feinman, 1983, p. 19). Female inmates were not considered "potential breadwrinners." The "aim to produce 'good housewrives' is clearly stated, and it is implied that successful family life depends on women's domestic skills" (Dobash, Dobash & Gutteridge, 1986, p. 64). 31 W Are these historical trends in any way reflected in the contemporary treatment of female prisoners? It is argued that the lack of understanding and biased ideology of women offenders is still prevalent today and "continues to intrude on the treatment of women in prison" (Feinman, 1979, p. 132). There appears to be an ideological shifting back to the ideal of "true womanhood": As evidence of this consider the new hostility signaled by the bringing of child abuse charges against women who use drugs even before the birth of their children (Chesney-Lind, 1991, p. 59). As argued above, female inmates have experienced various forms of social control through cultural stereotypes. This study focuses on three specific areas which possibly may have been influenced by cultural stereotypes: a) types of vocational programs for female inmates; b) motherhood and mothers in prison; and c) paternalistic attitudes toward female inmates. Mariam Pollock-Byme (1990) outlined programming for female inmates into five general categories: a) maintenance of the institution; b) educational; c) vocational; d) rehabilitative; and 9) medical care. In reference to vocational programs, Pollock-Byme (1990) conceded that [o]rdinarily, women's institutions do not have the same number or kind of vocational programs as are offered at institutions for men. For years, the only vocational programs available were those that prepared women for domestic service, clerical work, or cosmetology. Although nothing is wrong with such programs, and 32 they continue to exist at a number of institutions, many women have no interest in these fields or will need more lucrative employment upon release to support themselves and their children adequately (p. 91). Some have argued that vocational programming for female inmates reinforce traditional roles of women (Carlen, 1982; Carp & Schade, 1993; Chapman, 1980; Moyer, 1984; Simon & Landis, 1991). In a study of state-run facilities for women, Weisheit (1984) reported that most of the thirty-six institutions surveyed provided programming which reinforce the traditional roles of women. This programming included food services (N=28), secretarial (N=31), domestic work (N=20), and cosmetology (N=16). However, some institutions did offer some non-traditional programming: auto repair (N=8), carpentry (N=15), computer-related (N=21), electrical (N=15), and plumbing (N=12) (p. 37). Simon and Landis (1991) also conducted a survey of forty state-run institutions. This study revealed that vocational programs for women had a tendency to reinforce traditional roles: clerical/office skills (N=28); typing (N=25); data processing (N=19); cooking/domestic skills (N=17); food service (N=16); and cosmetology (N=16). As with Weisheit's study, Simon and Landis (1991) reported that there were also some non-traditional programs offered such as computers (N=10); building maintenance (N=10); carpentry (N=9); plumbing (N=9); graphics/painting (N=8); welding (N=7); and masonry (N=6) (p. 92). 33 In their article on programming for women and men in US. prisons, Morash, Haarr, and Rucker (1994) argued that there were gender disparities in work assignments. Women prisoners were over-represented in janitorial and kitchen assignments, but compared to male inmates, they were under- represented in farm and forestry, maintenance, and repair duties (p. 204-206). One approach to remedy the lack of vocational programming for female inmates has been through prisoner litigation. During the 19803 there was a relative increase in litigation from women prisoners (Herbert, 1985). One well known decision for parity of treatment was Wham, 478 F .Supp 1075 «ED. Mich.) 1979)). The plaintiffs claimed that the vocational programs offered to female inmates was inferior to male inmates: The State's failure to ascertain the interests and needs of its female inmates, they claim, has resulted in a set of programs which prepare the participants for low-paying menial positions in fields traditionally occupied by women (p. 1086). In its ruling, the court ordered parity of programs in female institutions. Schweber and Feinman (1985) concluded from their study on legislative action for women's prisons that "the political reality of equality for women is that court orders and statutes are often just the beginning of the struggle" (p. 9). Litigation efforts, such as filgyer, may be "just the beginning" in an attempt to provide equal opportunities in vocational programming for female inmates. There are two essential aspects concerning those vocational programs that reinforce traditional roles of female inmates. One aspect involves the 34 choice of occupations considered to be appropriate for women. Occupations considered as appropriate for women are those fields which traditionally have been held by women (e.g., clerical and food service). To examine this facet of vocational programming, this study explored the following research question: Research Question: Is group membership related to attitudes about traditional and non-traditional work for female inmates? Another aspect of vocational programming involves presumptions that women will be able to financially depend on men, particularly their husbands. Thus, women do not need to obtain employment that will provide them with financial stability and independence. However, as cited above, "women will need more lucrative employment upon release to support themselves and their children adequately." Female inmates' economic reality is that they are usually the primary caretaker for themselves as well as their children. To explore this facet of vocational programming, this research addressed the following research question: Research Question: Is group membership related to attitudes about a female inmate's need to work? In Figure 1, these two research questions are depicted by the line connecting Group Membership and the box labeled "Vocational Programming," with Social Distance as a controlling factor. There are various implications if 35 group membership is related to attitudes about vocational programming. For instance, if such stereotyping is supported by program staff and correctional officers, then there are implications for reuniting, training, and supervision necessary to promote non-traditional programming. Additionally, if such stereotyping is supported by female inmates, then there are implications and possible limitations for their participation in non—traditional vocational programs. Hamming Historically, societal attitudes toward the parental rights of female offenders differed from those of male offenders (Beckerrnan, 1991): The female felon offends society's idealized vision of women as all- caring, nurturing, and attentive to their children. She therefore poses a threat to the established social order unlike that presumably posed by male felons. Thefemale felon's criminal activities raise concerns about her ability to be a "good" mother (p. 1 72). Inherent in this idealization of motherhood is that women have a natural, maternal instinct: This imposition of moral assumptions and expectations, and their translation into 'natural' laws, is a forceful way of maintaining the female image. Thus, the 'matemal instinct' is natural and it is 'natural' for women to assume a caring role. Conversely, it is 'unnatural' for a women not to want children and it 'goes against her nature' to seek an abortion (Amir 8. Biniamin, 1991, p. 22-23). These stereotypes imply that a man, as a parent, is replaceable (e.g., stepfather). However, a mother, due to her "maternal instinct," is not replaceable. 36 In 1991, the Bureau of Justice Statistics reported that over three-quarters of the women had children with an estimated 25,700 female inmates having more than 56,000 children under the age of 18. While incarceration is the first significant separation for many mothers and children, the mother and child separation can occur in various situations such as hospitalization, school, or military service. Separation due to incarceration, however, acquires an additional imputation (Baunach, 1985). Not only is a mother separated from her child, but that separation is due to her "deviant" behavior or criminal activity. Furthermore, separation for an incarcerated mother may differ from that of an incarcerated father due to the fact that in most instances the children were not living with the father prior to incarceration (Datesman 8. Cales, 1983). In her study on mothers who use crack cocaine, Maher (1992) argued that in Western cultures, there is a romanticized, idealized notion of the family which is constructed on the regulation and control of women (p. 39). There are idealized constmctions of what constitutes a "good" and a "bad" mother. Citing Davin (1978), Maher (1992) illustrated the pervasiveness of the "motherhood" ideology: Motherhood was so powerful a symbol that often class differences disappeared, along with the realities of working class life. All the individual real mothers were subsumed into one ideal figure, the Queen Bee, protected and fertile, producing the next generation for the good of the hive. . . . The family was such an accepted symbol for the state that its actual disparate identities were forgotten (p. 40). 37 Mahan ( 1982) provided a critical examination of motherhood and incarceration by providing narratives from eleven women incarcerated in a New Mexico state correctional facility. One woman's experience of this idealization illustrates its pervasive and effectual influence: Separated from her child by time and space, Helen found her life farther and farther removed from the institution of motherhood. She described a painful conversation she once had with the prison supervisor about her child. He had asked her, "If you care about your kid so much, why are you a bad mother?" She could not answer. She saw the role of mother as lost to her. . . . In this way Helen remained outside the institution of motherhood but looked to her own behavior for the loss. She never questioned the institution itself (p. 1 19-120). This ideology is constructed within patriarchal and Judeo-Christian systems of beliefs. The perception of women inmates as bad mothers was elucidated by Mann (1984) when discussing pregnant inmates: a.) [ljt has been observed that an incarcerated pregnant women may indeed be treated worse than other inmates because of her condition (p. 227). Someoffithecorrectional staff may harass pregnant/inmates and display hostility or disapproval toward them. This perception of a "bad" mother can also be internalized by the female inmates themselves. Prior to imprisonment, some women may have been involved in behaviors that would have led to state intervention such as substance abuse, abandonment, or abuse (Pollock-Byme, 1991). As cited by Pollock-Byrne (1991), Henriques' (1982) study indicated that 38 in these instances the mother believed that the primary caretakers responsible for her children were relatives. However, it does seem to be true that some women in prison often abdicated their responsibility to other family members before being incarcerated, this failure only reinforced their views that they were "bad" mothers who let their children suffer. Such a view often influences feelings of depression and powerlessness in prison, and contributes to fantasies about what life will be like outside after release (p. 67). This perception could also be further perpetuated by family members. Baunach (1985) noted the limited emotional support from the prison for maintaining ties between incarcerated women and their children. Citing Lundberg, Sheckley, and Vuelkar's (1975) study, such prison support systems are minimal with respect to encouraging female inmates to express grief as well as coping with separation. This lack of support is further exacerbated when the prison system encourages dependency in inmate mothers: Lundberg et al. noted that prisons create a forced dependency that is "antithetical to the requirements of the mother role." Inmate- mothers become dependent upon the institution for survival and are unable to take responsibility for themselves or for their children. Therefore, it should not be surprising that they cannot resume care of their children upon release (p. 8). One suggested approach to strengthening the mother-child relationship has been to provide programs within the institution (Baunach, 1985; Datesman & Cales, 1983; Lundberg, Sheckley, 8. Vuelkar, 1975). There have been various programs offered to establish as well as maintain the relationships between 39 female inmates and their children. (Baunach, 1979; Datesman & Cales, 1983; Hale, 1977; Mann, 1984; McCarthy, 1980; Neto & Ranier, 1983). LeFlore and Holston (1989) compared attitudes about parenting behaviors between mothers in prison and mothers on the "outside." From their findings, they agreed that mothers in prison should be provided the opportunity to maintain ties with their children. Furthermore, "she should be supported in her efforts to fulfill her social role of mother and be encouraged to integrate this with her ability to 'parent at a distance'" (p. 19). The literature points to two factors relevant to stereotyping and mothers in prison. One factor is the idealization of motherhood. This idealization results in the perception of mothers in prison as "bad" mothers. This perception is not only due to separation from their children, but that this separation is caused by their criminal activity. To examine this aspect of mothers in prison and motherhood, this study addressed the following research question: Research Question: Is group membership related to attitudes about inmate mothers being "bad" mothers? Another aspect of motherhood and mothers in prison involves maintaining ties with their children. While mothers are separated from their children due to incarceration, Baunach (1985) argued that there is limited support within most prisons to maintain these ties. To explore attitudes about maintaining ties 40 between mothers in prison and their children, this research examined the following research question: Research Question: Is group membership related to attitudes about maintaining ties between inmate mothers and their children? In Figure 1, these two research questions are depicted by the line connecting Group Membership and the box labeled "Parenting," with Social Distance as a controlling factor. Efforts to implement programs and policies for inmate mothers is characteristically an institutional level approach. This research focused on a more individual level of study. As discussed previously, the various groups within the institution interact with the inmates on a day-to-day basis. Therefore, one questions, "If these individuals hold certain stereotypes of inmate mothers, will these influence their interactions with the women?" Eatemallsm Paternalism has been interpreted as follows: The derivation of the term "paternalism" from a Latin-English kinship term suggests its root meaning: a type of behavior by a superior toward an inferior resembling that of a male parent to his child. . . Within different types of paternalistic systems, the following three basic ideas . . . can be found. First, since a "child is defenseless and lacks property, he requires assistance and support. Second, since a "child" is not fully aware of his role and therefore not fully responsible, he requires guidance. . . . The third 41 idea holds that since a "child" is ignorant, he can be deceived, or treated in such a way as to serve the interests of the "adult," without becoming aware of this (International Encyclopedia of the Social Sciences, 1968, p. 422). Historically, women inmates were stereotyped as being "childlike" or "infantile." This paternalistic approach towards female inmates was a guiding correctional management strategy (Sargent, 1984, p. 41). Freedman (1981) noted that at the Chrittenton Homes, regulations stipulated that "[i]nmates when admitted are adopted into the family and are expected to be given the loving obedience of dutiful children towards their parents" (p. 56). Some researchers contend that this stereotype persists today: From the founding of the first homes for discharged prisoners, through the family-style reformatory systems, to present-day correctional institutions, women prisoners have been forced to play the parts of children. Just as some superintendents in the past called their charges "the girls," so later prison personnel have continued to view inmates "as being weak, like children," and have treated them accordingly (Freedman, 1981, p. 154). Some female inmates are often referred to as "girls" or "ladies" in a tone of voice reminiscent of a parent talking to an adolescent. Fox (1984) provided the following example as to how inmates perceive this type of treatment: For a 33-year-old women to have another woman tell her that she is misbehaving, it's funny to me, it's funny! I have a child who is 15 years old and I wouldn't tell him that he is misbehaving. He'd look at me like I was crazy. To be 33 and have somebody tell you that you're acting like a little child, and that means that you're going to be punished, it's funny. And if they do disrespect me, I'll call their attention to it. I'll say, "Excuse me, I would like to speak with you. From now on, if you feel that you cannot respect me or say anything nice to me, don't say anything at all." And if I want to do 42 anything about it, I'll just do it. And I take the consequences (p. 23). Burkhardts (1973) study on female prisons provided an interesting and illustrating example as to how these stereotypes are experienced by the inmates: "T he main thing I think kept me in Muncie so long was they was trying to rehabilitate me down to be a two-year-old. I refused to be retarded," she said. "I'm grown. I'm forty-nine years old. They can't make me be a lwo-year-old' (p. 128). She had often heard that women inmates were "babied" when compared to man. What she found perplexing was that she did not observe this type of "coddling" behavior but continued to hear these assertions. For example, Lieutenant Archibald at Rikers Island, New York City: "They're grown up, but they act like they're in kindergarten." Pointing to one forty-year-old woman who was quarreling with another inmate over a seat in the auditorium, the lieutenant said, "She's my problem child" (p. 127). Burkhardt further noted that some administrators would make such comments about their "feedings" and "controlling the girls" and that the women were "acting just like babies" (p. 128). The literature suggests that female inmates are perceived as more emotional and excitable compared to male inmates. Thus, female inmates need more emotional support. In her study on correctional officers' stereotypes of female inmates, Pollock (1984) reported that eighty-nine percent of the officers agreed with the statement that female inmates were more emotional than male inmates. Below are two examples from this study that elucidate the officers' perceptions of the female inmates' "emotionality": 43 They'll become hysterical, saying you're destroying their dignity or some such thing in a strip search, a male will accept that more readily I think, although they don't particularly care for it either. . . . (female correctional officer) I think they lose their temper much easier, fly off the handle more easily, they have less control of themselves in any given situation, they get extremely upset extremely quick (male correctional officer) (p. 87). Most female inmates may be more likely to express their emotions. However, what becomes problematic is when this open expression of emotions is devalued or deemed as a negative trait: [Mjen seem to be more stable than women. One woman can be at a high one day and the next day she can be very low, and the men generally you get one man and he's generally that way most of the time unless something comes up, you know, to push him over the edge. But generally, they're the same way the whole time you know them. The women go up and down (Pollock, 1984, p. 87). In reference to interactions between female inmates and correctional officers, Pollock-Bryne (1991) noted that the perception of the female inmate as needing more emotional support and guidance in establishing her femininity is widespread in the literature concerning not only females but also adult women offenders. This view of female inmates creates an environment where women ofi‘enders are treated very often as children; they are called “girls“ or ladies, " and the tone is alien that used to discipline teenagers, or somewhat dense and naughty adults [italics added] (p. 122). Two aspects of paternalistic attitudes toward female inmates have been revealed in the literature. One facet is the perception that female inmates are irrational. The following research question explored if such stereotypes exist among the various groups within the institution: 44 Research Question: Is group membership related to attitudes about female inmates being irrational? Another aspect of paternalistic attitudes toward female inmates is the perception that they act like children. The following research question explored this facet of paternalistic attitudes: Research Question: Is group membership related to attitudes about female inmates being child-like? In Figure 1, these two research questions are depicted by the line connecting Group Membership and the box labeled "Paternalism," with Social Distance as a controlling factor. ATTITUDE-BEHAVIOR RELATION As stated above, not only has the criminological literature claimed the existence of stereotypes, but arguments have been made that such attitudes towrards female inmates negatively influences their lives. These arguments imply behavioral consequences of such attitudes. Generally, however, in the literature on attitudes and behaviors, there are questions raised concerning this relation. 45 Attitude-Behavior Consistency Allport (1935) conceptualized an attitude as a "mental or neural state of readiness, organized through experience, exerting a directive or dynamic influence upon the individual's response to all objects and situations with which it is related" [italics added] (p. 810). This definition implies a strong relation between an individual's attitude and her or his behavior. However, the one-to- one relation between attitudes and behavior has seriously been challenged. One of the earliest challenges to this consistency concept was I.aPiere's (1975) study on hospitality toward Chinese. In the late 19603, Wicker (1969) conducted a comprehensive review of thirty-one studies examining the strength of the attitude-behavior relation. He concluded that "[t]aken as a whole, these studies suggest that it is considerably more likely that attitudes will be unrelated or only slightly related to overt behaviors than that attitudes will be closely related to actions" (p. 65). Rather than abandoning the theoretical proposition that attitudes and behavior are related, researchers have attempted to provoke further inquiry and to improve understanding of this relation (Abelson, 1982; Liska, 1974, 1975; Ronis, Yates, 8. Kirscht, 1982; Sherman 8. Fazio, 1983; Shuman & Johnson, 1976). Ronis, et al. (1982) categorized various approaches that have been suggested to improve the attitude-behavior relation. This research draws on the 46 methodological and the multivariate or "other" approach to understanding the relation of attitudes to behavior. Methodological Approach The methodological approach to improving the attitude-behavior relation has been to identify problems of measurement. A great deal of work has been to focus on the generality and specificity of the attitudinal measure as well as the implementation of a single-act or multiple-act criterions of behavior (Ajzen, 1982; Weigel 8. Newman, 1976). Ajzen (1982) argued that global attitudinal measures are not strong predictors of single actions. Instead, implementing multiple-act measures of behavior improve the correlation between attitudes and behaviors. Ajzen cited Weigel and Newman's (1976) study of protecting the environment to illustrate the strength of multiple-act measures. To measure attitudes toward protecting environmental quality, the researchers administered a sixteen item Likert scale. To measure behaviors to protect the environment, fourteen behavioral observations were made including participating in a litter pick-up program and a recycling program. Weigel and Newman developed four multiple-act indices: petition-signing behaviors; litter pick-ups; recycling; and an overall index based on all fourteen single behaviors. Prediction of single act measures from the attitudinal measure was weak with an average correlation of .29 and not significant. However, the average correlation with the three behavioral indices 47 was .42 and significant. The correlation with the index based on all fourteen behaviors was .62 and also significant. Ajzen (1982) argued that empirical evidence has demonstrated that general measures of attitudes are poor predictors of single actions. The low correlation between attitudes and behaviors can be improved by selecting the appropriate specificity and generality of measures. From the empirical evidence, two conclusions were offered: A) Global attitudinal measures are appropriate for the prediction of global behavioral tendencies rather than single actions. B) Single actions are best predicted from specific attitudinal measures (p.11). Multivariate or "Other" Approach The multivariate research approach focuses on identifying factors, or moderating variables, that can improve the attitude-behavior relation. As Ronis, et al. (1981) argued, "[i]f attitude were the only determinant of a behavior, attitude and behavior could correlate perfectly. This is not possible if other factors affect the behavior" (p. 215). Such moderating variables include temporal instability (Schwartz, 1978); attitude structure (Ajzen, 1989; Bagozzi 8. Bumkrant, 1979) as well as "social distance" and "social constraint" (Warner 8. DeFleur, 1969). 48 One moderating variable in F ishbein and Ajzen's (1975, 1981) theory of reasoned action was behavioral intention. They argued that the immediate determinant of an individual's overt behavior is her or his intention to perform or not perform that behavior. So, if you knew what a person's intention was regarding some object or person (e.g., does the person intend to vote for Candidate A or B; does the person intend to purchase Brand X or Brand Y), this would be the single most important piece of information that you could have in attempting to predict the person's eventual behavior (Petty & Cacioppo, 1981, p. 193). Another aspect of the present study examined the relation between the specific attitudes of female inmates as discussed above and the possible behavioral intentions to interact with female inmates in various situations. To assess if attitudes toward vocational programming are related to interactions between the various groups, the following research questions were explored: Research Question: Are attitudes about traditional and non-traditional work for female inmates related to behavioral intentions to encourage or discourage an inmate to participate in non-traditional programming? Research Question: Are attitudes about a female inmate's need to work related to behavioral intentions to encourage or discourage her to be financially independent? Figure 1 illustrates the relations to be examined between attitudes concerning vocational programs and the intended behavioral interactions: a) encourage 49 female inmates to participate in traditional or non-traditional fields; and b) encourage female inmates to be financially independent. To determine whether stereotypes concerning mothers in prison could influence behavioral interactions with inmates, the following two research questions were addressed: Research Question: Are attitudes about mothers in prison being "bad" mothers related to behavioral intentions to encourage mothers to retain custody of their children? Research Question: Are attitudes about inmate mothers' maintaining ties with their children related to behavioral intentions to encourage them to maintain ties with their children? The relation between attitudes toward inmate mothers and behavioral intentions is depicted in Figure 1. The behavioral intentions are as follows: a) encouraging inmate mothers to retain custody of their children; and b) encouraging inmate mothers to maintain ties with their children. Again, two research questions explored the relation between paternalistic attitudes and intended behavioral interactions with female inmates: Research Question: Are attitudes about female inmates being irrational related to behavioral intentions to interact with inmates in a stressful situation? Research Question: Are attitudes about female inmates being child-like related to behavioral intentions to treat inmates like children? 50 Again, these research questions are illustrated in Figure 1. The two behavioral intentions are as follows: a) interactions with an inmate in a stressful situation; and b) treating inmates as children. For the above six research questions, the relation to be examined is exploratory rather than explanatory. A final component of this research examined if an individual's attitude toward women can explain a portion of the relation between the specific attitudinal measure and the behavioral intention measure. If the global measure of attitudes toward women explained a significant amount of the relation, there are notable implications for training and recruiting. For instance, to improve interactions between the various groups within an institution, one needs to ascertain if the focus should be on an individual's specific attitude toward female inmates or an individual's global attitudes toward women. To further explore the relation between the attitudes toward female inmates, attitudes toward women, and behavioral intentions, the following research questions were addressed: Research Question: How much of the variation within the behavioral situation of encouraging an inmate to participate in non-traditional programming can be explained by attitudes about traditional and non-traditional work for women and how much of the variation can be explained by global attitudes toward women? Research Question: Research Question: Research Question: Research Question: Research Question: 51 How much of the variation within the behavioral situation of encouraging a female inmate to be financially independent can be explained by attitudes about a female inmate's need to work and how much of the variation can be explained by global attitudes toward women? How much of the variation wn'thin the behavioral situation of encouraging inmate mothers to retain custody of their children can be explained by attitudes about perceptions of inmate mothers being "bad" mothers and how much of the variation can be explained by global attitudes toward women? How much of the variation within the behavioral situation of encouraging inmate mothers to maintain ties with their children can be explained by attitudes about inmate mothers separation from their children and how much of the variation can be explained by global attitudes toward women? How much of the variation within the behavioral situation involving interactions with inmates in a stressful situation can be explained by stereotypes about female inmates being irrational and how much of the variation can be explained by global attitudes toward women? How much of the variation within the behavioral situation involving the treatment of inmates like children can be explained by stereotypes about female inmates being child-like and how much of the variation can be explained by global attitudes toward women? 52 PILOT STU DY A pilot study was conducted in a Michigan county jail facility in an urban loeation. This facility housed both male and female inmates. The sample consisted of ten correctional officers. The purpose of this pilot study was two- fold. First, the pilot study enabled the researcher to acquire an approximate time-frame as to how long it would take the participants to complete the questionnaire. It took the participants approximately 25 to 30 minutes to complete the questionnaire. This information was helpful when conducting the study at the women's correctional facility. Second, the pilot study provided the researcher with feedback on questions that were obscure or unclear to the participants. After completing the questionnaire, a few participants commented that the instructions pertaining to the social distance scale were obscure. Therefore, these instructions were further clarified with an example provided at the beginning of the questionnaire as to how the participants should respond to this instrument. BEDFORD HILLS CORRECTIONAL FACILITY The setting for this study is the Bedford Hills Correctional Facility in the southern part of New York State. Because this study has been conducted in one specific correctional facility, it is essential to provide a description of the Bedford Hills Correctional Facility (Bedford Hills Correctional Facility, 1994). 53 II'IIIQ . The Bedford Hills Correctional Facility was established in 1892 under the name of the New York State Refonnatory for Women. A prominent proponent for the establishment of the Refonnatory was Abby Hopper Gibbons, who was head of the New York Committee for the Suppression of Legalized Vice (Rafter, 1 990, p. 47). The guiding philosophy of the institution was to reform the offenders to become proper women in society. The architecture was designed to reflect this philosophy by constructing cottages in order to provide a "home-like" setting. The Refonnatory opened in 1901. It was initially intended for women convicted of misdemeanors as well as property felonies. The Refonnatory also housed juvenile delinquents and "wayward minors." Due to government reorganization in 1926, the Refonnatory became part of the Department of Corrections. In 1932, the New York State Refonnatory was then named the Westfield State Farm which was eventually renamed the Bedford Hills Correctional Facility. B "5 lllill E l' IE 1'! I I The Bedford Hills Correctional Facility is designated as a maximum security correctional facility for females. It is the only maximum security facility 54 for women in the State of New York. The facility consists of three separate components: A) A reception and classification center for all females entering the Department's custody, and the HUB (Orientation) Assessment Unit for females. B) A detention center for female parolees or conditional release violators with hearings pending. C) A general confinement facility for female inmates 16 years of age or older (no). The present study focuses on those individuals who primarily fall within the last component (i.e., general confinement). The mission statement for this component of the facility is as follows: Bedford Hills Correctional Facility houses . . . inmates assigned to its general population within a secure setting in which to carry out court sentences and insure the protection of the community by the: A Prevention of escapes B. Creation and maintenance of a safe, secure, and humane environment which is protective of the life, health, and safety of the inmates, staff and the community C. Maintenance of a wide range of habilitative programs relevant to the reintegration of women inmates into the community D. Development of pride of purpose through employee participation in training and involvement of employees in staff meetings and on committees which foster improved communications between organized labor and management (n.p.). At the time of this study there were approximately 775 inmates housed at Bedford Hills. Women can be housed in the general confinement component of Bedford Hills for any security or medial classification. 55 Bedford Hills offers various programs to these female inmates. Services provided within the counseling unit include the following: Individual/Group Counseling; Hispanic Inmate Needs Coordinator, Alcohol/Substance Abuse Treatment Program (ASAT); Down on Violence/Altematives to Violence Programs; and, Crisis Intervention. Services are also provided for religious and family needs. The educational programs included ABE Reading/Math; Pre GED; GED; English as a Second Language; Bilingual Education; Computer Lab; Education Outreach; Literacy Tutoring; Health Science; and Library. The various occupational or vocational programs included Business Education; Building Maintenance; Cosmetology; Printing; Horticulture; and the Department of Labor Apprenticeship Program. This last program provides inmates the opportunity to obtain certified apprenticeships through the printing and horticulture classes. The Mother's and Children's Program was designed a) to assist mothers and their children to maintain ties; and b) to provide mothers instruction, if needed, on parenting. It consists of various components such as The Children's Center", Foster Care Committee; Parenting through Films; Intensive Parenting Program; Summer ProgramNVeekend Program; Christmas Program; and Choices and Change. There is also a Nursery Unit which houses approximately 26 inmates and their babies. The facility also offers the Family Violence Program which addresses women's experiences of victimization (e.g., battered Women; survivors of child abuse; and incest survivors). 56 There are also organizations which the inmates themselves help operate. These include ACE (AIDS Counseling and Education); Long Terrner's Committee; Women Helping Other Women (WHOVV); National Organization for Women (NOW); Facility Photography Program; Sunni-Muslim Literary Guild; and Hispanic Needs Committee. As illustrated by the above discussion, Bedford Hills Correctional Facility is a distinctive institution. It provides various program opportunities for female inmates. Another unique feature of Bedford Hills is the training and certification program for female inmates to eventually become peer counselors. Specifically, this program trains female inmates to deliver various services (e.g., tutoring, family education) to other inmates in the institution. Again, this is a special group within this institution. In fact, the existence of peer counselors is unusual in most female correctional facilities across the country. Therefore, conducting this study at Bedford Hills also provided an opportunity to survey this unique group of peer counselors and, subsequently, provide another comparison group for the present study. 57 ENDNOTES Interestingly, among all the inmates surveyed, 25% of the mothers and 90% of the fathers, reported that their children were living with the other parent while incarcerated. Therefore, 75% of inmate mothers' children were cared for by non-parental adults. CHAPTER THREE RESEARCH METHODS A survey method was utilized for this study. The survey instrument was administered to different groups at the Bedford Hills Correctional Facility. The survey was based on a cross-sectional design. This method was the most appropriate for the present research, especially since it entailed studying various groups within the institution (Babble, 1990; Rea 8. Parker, 1992). Rather than one-on-one interviews, the administration of the surveys was conducted with a relatively large number of participants. Sample Data were collected from four samples: a) female inmates; b) female inmates who are peer counselors; c) correctional officers; and d) program staff. The definitions for each group are as follows: a) Female inmates are those women who have been convicted of a felony and incarcerated in the general population at Bedford Hills Correctional Facility. b) A female inmate peer counselor is an individual who meets the definition of an inmate but also has responsibilities that include some type of program delivery duties within the institution. 58 59 c) A correctional officer is an individual who is employed at Bedford Hills Correctional Facility as a correctional officer. d) Program staff are individuals who are employed in the institution to provide program services, such as education and counseling, to the female inmates. Initially, the female inmate sample was designed to be selected by using a random sampling method. A current population sheet of the entire inmate population was obtained. The population sheet was dated 17 April 1996.1 There were 775 inmates housed in the facility at that time. From this list, the entire inmate population was numbered consecutively. Thereafter, by using a table of random numbers, various inmates were selected for the study. Two staff personnel checked these names to ensure that any individuals on the list designated with a mental health problem were removed. The remaining individuals were then placed on a "call-out" sheet for the following day to meet with the researcher. The researcher then asked if they would like to participate in the study. Those few individuals who did agree to participate then completed the survey. However, this approach did not result in a large number of participants. Approximately fifteen inmates had participated during these testing sessions. Many inmates had a conflict in their schedule between participating in the study or attending class and/or work. Furthermore, attending class or work was considered a higher administrative priority than participating in a study. 60 Therefore, the sampling method had to be modified in order to acquire more female inmates in the sample. A larger number of surveys were then distributed in the classrooms. From the one hundred surveys that were distributed, a total of 74 were returned, resulting in a 74.0% response rate. The researcher obtained a list of peer counselors from the prison administration. There were a total of 57 inmates listed as peer counselors. Initially, the same random sampling method was to be implemented for this group as well. However, the same problems in acquiring the female inmates also affected this group. Therefore, many of the peer counselors were approached while they were involved with their counseling duties. They were then asked to participate in the survey. Of the 50 surveys distributed, 29 completed the questionnaires resulting in a 58.0% response rate. Obtaining the sample of correctional officers was the most problematic. Initially, a list of the officers was obtained from the administration. This list included the correctional officers' names and their date of employment. The list was then separated into "new/' and "senior" correctional officers. The "new" correctional officers were designated as any individual who was recently employed at Bedford Hills (i.e., two years or less). There were a total of 141 "new" correctional officers. A "senior" correctional officer was designated as any individual who had been employed at Bedford Hills for at least more than two years. There were a total of 229 "senior" correctional officers. For each of these separate lists, the individuals were numbered consecutively and were 61 subsequently to be randomly selected using a table of random numbers. However, the researcher later Ieamed that surveying these individuals with this approach is extremely difficult. Many correctional officers take extra shifts in order to acquire more consecutive days off. Therefore, some of those officers selected for participation would not be at work during the study. Furthermore, it would be difficult for the researcher to survey those officers during the evening shift. In order to do so, a counselor would have to escort the researcher during the evening to distribute these surveys. Most counselors, however, did not work during the evenings. Prior to beginning their shift, correctional officers are required to attend "line-up" which is conducted by a lieutenant. The lieutenant addresses various issues pertaining to their duties. The researcher obtained perrnlssion from the lieutenant to distribute the surveys during "line-up." The majority of surveys were distributed in this manner. Additional surveys were distributed to correctional officers while on their various posts. Only 16 correctional officers returned their surveys. One hundred surveys were distributed, resulting in a 16% response rate. During the distribution and collection of the surveys, the researcher was informed that many correctional officers were not completing the surveys primarily due to either a perceived threat to confidentiality or the length of time involved in completing the survey. In an attempt to increase this number, fifty additional surveys were mailed to a contact person at Bedford Hills. These surveys were to be distributed 62 during another "line-up." From these fifty, an additional five were completed resulting in a total number of 21 correctional officers. The researcher was informed that the same reasons listed above were also offered as to why officers were not completing the surveys. Therefore, the "new" and "senior" correctional officer groups were combined into one group (i.e., "correctional officers").2 Again, the sample of program staff was to be obtained in a similar method as the correctional officers. However, this was also problematic essentially due to the relatively small number of program staff. There were approximately 65 program staff including reception counselors. Reception counselors, however, were not included in this sampling group. Therefore, the surveys were distributed to various program staff while they were on their jobs. From a total of fifty surveys, 27 were returned resulting in a 46.6% response rate. Table 1 is a summary of the sample groups for this study. Table 1 Summary of Sample Groups Female Inmates Peer Counselors Correctional Officers 63 Survey Instrument The survey instrument consisted of four major sections: a) background information; b) a social distance measure; c) attitudinal measures; and d) behavioral intention measures. All the questionnaire items were In a closed- ended response pattern. The consent form was the first page of each questionnaire. After the respondent signed this form, she or he was assigned a case number. The consent form was then removed from the questionnaire and placed in a separate location in order to ensure the participant of confidentiality (see Appendix A). Backemundjntennation The first major portion of the survey asked the respondents for some general demographic information such as age, race/ethnicity, gender, educational level, etc. lnforrnation pertaining specifically to each group was also included such as current offense serving at Bedford Hills (e.g., female inmates and peer counselors), peer counseling responsibilities, services provided to inmates (e.g., program staff), and number of years employed at Bedford Hills (e.g., program staff and correctional officers). (see Appendices B through E). 54 SociaLQIstance The second section measured the social distance among the various groups. Social distance was operationalized as interactions between organizational groups within the prison setting rather than ethnic or racial groups. The scale, "Social Distance," consisted of seven items (see Appendix F). The items ranged in various interactions from "avoiding" to "talking about personal problems" to the five groups within this study (i.e., female inmates, peer counselors, new correctional officers, senior correctional officers, and program staff). If a respondent would engage in this type of behavior, that item was coded "1." If a respondent would not engage in this type of behavior, that item was coded "0." To develop the Guttman scale format, the item 'Would you avoid . . . " was subsequently deleted from the scale. If a respondent was to avoid an individual from a certain group, that would be reflected by she or he not checking any of the interactional items in the scale. The scores ranged from "0" to "6." A "0" indicates that the respondent would have no interaction with a member from that group while a "6" would indicate that a respondent would have a significant interaction with a member from that group (i.e., discuss personal problems). To assess the reliability of this scale, a coefficient of reproducibility was calculated. To assess the effectiveness of the reproducibility coefficient, a coefficient of scalability was calculated. An acceptable coefficient of 65 reproducibility is between .90 to .95 (Babble, 1986; Murphy & Tanenhaus, 1972). An acceptable coefficient of scalability is .60 (Dunn-Rankin, 1983; Murphy & Tanenhaus, 1972). Two coefficients of reproducibility were calculated for this scale (.95 and .87). They differed depending on the method used to score mixed types (Babbie, 1986).3 The coefficient of scalability was .63. This falls within the range for an acceptable Guttman scale. William The second major portion of the survey consists of the various attitudinal measures. This portion consists of one global measures of stereotypes toward women and then seven specific measures of cultural stereotypes of female inmates: a) Attitudes about Non-Traditional Work for Female Inmates; b) Female Inmates Need to Work Scale; c) Maintaining Ties with Children; d) ldealizations of Motherhood; e) Child Visitation Scale; f) Paternalistic Attitudes Towards Female Inmates; and 9) Female Inmates are Irrational. Corresponding scales for male inmates were also constructed in order to assess if there were differences in attitudes concerning female inmates and male inmates. We The global measure of stereotypes is the Sexist Attitudes Toward Women Scale developed by Benson and Vincent (1980) (see Appendix G). The Sexist Attitudes Toward Witnesses/e senses of seven dimensions: 66 1. Attitudes that women are genetically inferior (biologically, emotionally, intellectually) to men. 2. Support for the premise that men should have greater rights and power than women. 3. Support for sex discrimination (antifemale) practices in education, work, and politics. 4. Hostility toward women who engage in traditionally masculine roles and behaviors or who fail to fulfill traditional female roles. 5. Lack of support and empathy for women's liberation movements and the issues involved in such movements. 6. Utilization of derogatory labels and restrictive stereotypes in * describing women. 7. Evaluation of women on the basis of physical attractiveness information and willingness to treat women as sexual objects (p. 278). Benson and Vincent claimed that although some scales have focused on one or more of these dimensions, no scale has encompassed the multidimensional characteristics of sexism: [The Sexist Attitudes Toward Women Scale] is a better measurement of the sexism construct than scales that measure only one or two of the hypothesized components of sexism (e.g., sex-role scales) (p. 287). The researchers reported a coefficient alpha of .93 for nonstudent adults and .90 for college students. The scores range from 40 to 280, with a higher score indicating greater sexist attitudes. The coefficient alpha for this research was .860 for the entire sample. Dugger (1991) implemented a scale to compare gender-role attitudes of Black and White women. The scale consisted of eight dimensions. One 67 dimension, public sphere, measured the respondents' attitudes about a woman's ability to perform competently in public-sphere positions that were predominantly male. The item read as follows: Women are entering all kinds of different fields of work these days. I'm going to name some different occupations. In each case, and assuming that you didn't know the person, would you tell me if you would have more confidence in a man in that situation (coded 1), or more confidence in a woman, or wouldn't it have any effect on your confidence whether it was a man or a woman (the latter two response categories were both coded 2). Would you have more confidence in a man or a woman as --— (p. 55). The scale items focused specifically on women in occupations within the public sphere that have traditionally been held by men. The reliability was .80 for both Black and White women. To measure encouragement of non-traditional occupations, a modified version of Duggei‘s scale was administered in this research. The items for this scale focused on various occupations outside the public sphere. Eleven items listed jobs characterized as being primarily for females and eleven items listed jobs characterized as being primarily for males. Encouraging either a male or a female inmate was coded 1. Encouraging both a male and a female was coded 2. The scores ranged from 22 to 44, with a higher score indicating encouragement of inmates to purse various occupations regardless of gender. Face validity of this scale is implied by studies comparing vocational programs for female inmates (e.g., Simon 8. Landis, 1991). For this sample, the scale had a coefficient alpha of .963 (see Appendix H). 68 The remaining attitudinal measures were combined into one test. It consisted of 37 items that examined three general areas of study. Seven items measured attitudes about female inmates' financial situation. Sixteen items measures attitudes about mothers in prison. The remaining fourteen items measured paternalistic attitudes (see Appendix I). As mentioned above, in order to make comparisons between attitudes towards male and female inmates, the same 37 items were utilized but referred to male inmates (see Appendix J). Below is a discussion on the scales developed from these items based on the three general areas of study. W In addition to the Attitudes about Non-Traditional Work for Female Inmates Scale, there were seven items measuring female inmates' financial reality or their need to work. These items were in a five-point Likert format. A factor analysis resulted in three factors.‘ The first factor consisted of three items which resulted in the Female Inmates' Need to Work Scale (see Table 2 for the individual test items and corresponding factor loadings).5 The coefficient alpha for this scale was .709. Using the same three items for male inmates, the coefficient alpha was .688. The Female Inmates' Need to Work Scale ranges from 3 to 15 with a higher score indicating a respondent's attitude that female inmates do not need to be financially independent. 69 The second and third factors consisted of two variables each. However, when testing for scale reliability, these factors produced coefficient alphas of only .301 and .195. Therefore, these factors were subsequently deleted from further analysis. Table 2 Factor Loadings for "Female Inmates Need to Work Scale" (N=145) Factor Test Item Loading Many women in prison can find someone, such as friends or family, to help them financially when released. Upon release, female inmates usually have the option of working outside the home or getting financial support so they can care for family members. Women in prison will have a significant other in their life that will hel them financiall when released from «rison. Wise There were sixteen items that addressed various issues pertaining to mothers in prison such as visitation, separation, and ldealizations of motherhood. Factor analysis generated four factors from these sixteen items. The first factor consisted of five items which resulted in the Maintaining Ties IMth Children Scale (see Table 3). The coefficient alpha for this scale was .836. Using the same five items for male inmates, the coefficient alpha was .852. The 70 Maintaining Tes With Children Scale scores range from 5 to 25 with a higher score indicating a respondent's attitude that female inmates should not maintain ties with their children. Table 3 Factor Loadings for "Maintaining Ties With Children Scale" (N=146) Factor Test Item Loading It is always important for children to visit their mothers in prison. Upon release, most female inmates will be good mothers to their children. It is important for a mother to be reunited with her children upon release. Maintaining contact between mothers in prison and their children helps them reunite upon release. Children need to see their mothers when their mothers are in orison. The second factor also consisted of five items which resulted in the ldealizations of Motherhood Scale (see Table 4). The coefficient alpha for this scale was .709. Using the same five items for male inmates, the coefficient alpha was .725. The ldealizations of Motherhood Scale scores range from 5 to 71 25 with a higher score indicating a more negative attitude about female inmates and motherhood. Table 4 Factor Loadings for "ldealizations of Motherhood" (N=142) Factor Loading Many mothers in prison have a hard time being good parents to their children. Even through a mother is in prison, it does not mean she is a bad parent. Parents who are incarcerated tend to have some problems being_good parents. Most female inmates were not very good mothers prior to bein- locked up. The vast majority of mothers in prison need to learn the basics of how to be an ad - . uate oarent. The third factor consisted of three items. However, the coefficient alpha for these items was .643 for attitudes pertaining to female inmates and .623 for attitudes pertaining to male inmates. Therefore, these items were deleted from further analysis. The fourth factor consisted of two items which resulted in the Child Visitation Scale (see Table 5). The coefficient alpha for this scale was .765. Using the same two items for male inmates, however, the coefficient alpha was 72 .647. The Child Visitation Scale scores range from 2 to 10 with a higher score indicating a more negative attitude about children visiting their mothers in prison. Table 5 Factor Loadings for "Child Visitation Scale" (N=141) Test Item Children should not be exposed to seeing their mothers in prison. Children should have just limited contact with their mothers the are incarcerated. W There were fourteen items that addressed paternalistic attitudes towards female inmates. Factor analysis generated four factors from these sixteen items. The first factor consisted of five items which resulted in the Paternalistic Attitudes Towards Female Inmates (see Table 6). The coefficient alpha for this scale was .740. Using the same five items for male inmates, the coefficient alpha was .734. The Paternalistic Attitudes Toward Female Inmates scores range from 5 to 25 with a higher score indicating more paternalistic attitudes toward female inmates. 73 The second factor consisted of four items which resulted in the Female lnmates Are Irrational Scale (see Table 7). The coefficient alpha for this scale was .674. Using the same four items for male inmates, the coefficient alpha was .672. The Female Inmates Are ln'ational Scale scores range from 4 to 20 with a higher score indicating that female inmates are irrational. Table 6 Factor Loadings for "Paternalistic Attitudes Toward Female Inmates" (N=141) Test Item Women in prison are usually too emotional. Women in prison are a lot like children. Some women in prison throw "temper tantrums." Women in prison have a tendency to overreact. Man women in orison com Iain too mch. The final two factors did not produce a high coefficient alpha (.507 and .583, respectively). Therefore, these items were deleted from any further analysis. Table 8 is a summary of the attitudinal measures and their corresponding reliabilities for the attitudes toward female and male inmates scales. 74 Table 7 Factor Loadings for "Female Inmates Are Irrational" (NI-144) Women in prison are usually rational. Women in prison can rationally express their anger. In most instances, when a conflict occurs between two female inmates, it is easy to calmly resolve the situation. Many women in prison can reasonably deal with disa . . . intrnent. Table 8 Summary of Attitudinal Measures and Coefficient Alpha l Scale . Female Inmates Male Inmates Sexist Attitudes Towards Women Scale .860 -— Attitudes About Non-Traditional Work for Female Inmates .963 —- Female Inmates Need to Work Scale .709 .688 . Maintaining Ties with Children .836 .852 ldealizations of Motherhood .709 .725 Child Visitation Scale .765 .647 Paternalistic Attitudes Towards Female Inmates .740 .734 Female Inmates are Irrational .674 .672 75 BI . III I' M The final portion of the survey consisted of six behavioral intention situations. For each of the six scenarios, factor analyses were implemented in order to determine if these scales were measuring one or more underlying factors. ScenarisLQne The first scenario measured the respondents' self-reported intention to encouraging a female inmate to participate in non-traditional programming (see Appendix K). A factor analysis was implemented to assess if this scale was measuring one or more than one underlying factor. The analyses resulted in one factor. Deleting one of the items from this scenario scale resulted in an alpha of .820. The scores range from 4 to 16 with a lower score indicating a respondent's tendency not to encourage such participation. W The second scenario measured a respondent's intention to encourage a female inmate to be financially independent (see Appendix L). A factor analysis resulted in three factors. The first factor consisted of three items: "Scenario Two - Financial Independence." The coefficient alpha was .799 (see Table 9). 76 The scores range from 3 to 12 with a low score indicating that a respondent would likely recommend a female inmate to be financially independent. The underlying concepts of the remaining two factors refer to children. The second factor consisted of two items: "Scenario Two - Work Inside the Home." The coefficient alpha was .664 (see Table 10). The scores range from 2 to 8 with a low score indicating that a respondent would discourage a female inmate to work inside the home. Table 9 Scenario Two - Financial Independence (N=144) Factor Test Item Loading How likely would you tell Andrea that it is important for her to be financially independent on her own income? How likely would you tell Andrea that she should pursue some type of occupation outside the home? How likely would you suggest that Andrea participate in a vocational program in case she needs to obtain em to ent inthe future? The third factor also consisted of two items: "Scenario Two - Work Outside the Home." The coefficient alpha was .666 (see Table 11). The scale also ranges from 2 to 8 with a low score indicating that a respondent would encourage a female inmate to work outside the home. 77 Table 10 Scenario Two — Work Inside the Home (NE-140) Factor Test Item Loading How likely would you encourage Andrea to stay at home and raise her children? How likely would you encourage Andrea to work in the home while her boyfriend provides financial support Table 1 1 Scenario Two - Work Outside the Home (N=143) Factor Test Item Loading How likely would you tell Andrea not to stay home and raise her children? How likely would you tell Andrea not to work in the home full time while her boyfriend provides financial sue . rt for the famil Scenariolhme Scenario Three assessed a respondent's intention to encourage mothers in prison to retain custody of their children. Factor analysis resulted in three factors. Subsequent reliability analyses revealed a low coefficient alpha for two of the three factors (.535 and .352, respectively). Therefore those factors and 78 the corresponding test items were excluded from further analysis. One factor did result in an acceptable coefficient alpha (i.e., .739). It consisted of two items (see Table 12). The scale ranges from 2 to 8 with a low score indicating a respondent's likelihood not to encourage a mother to retain custody of her children. mm Scenario Four measured a respondent's intention to encouraging an inmate mother to maintain ties with her children (see Appendix N). A factor analysis resulted in two factors. The first factor consisted of two items: "Scenario Four — Encourage Maintaining Ties." The coefficient alpha was .858 (see Table 13). The scale ranges from 2 to 8 with a low score indicating that a respondent is more apt to encourage an inmate to maintain ties with her children. Table 12 Scenario Three (N=143) Factor Loading Test Item How likely would you encourage Casssandra to let her mother have permanent custody? .823 How likely would you tell Cassandra that the children would "be better off" with her mother? - 79 Table 1 3 Scenario Four - Encourage Maintaining Ties (N=144) Factor Loading How likely would you tell Rachel that it is important to maintain contact with her children? How likely would you encourage Rachel to continue havin- contact with her children while in orison? The second factor initially consisted of four items: "Scenario Four - Discourage Maintaining Ties." To improve the coefficient alpha, two items were subsequently deleted resulting in an alpha of .857 (see Table 14). Table 14 Scenario Four — Discourage Maintaining Ties (NI-141) Factor Test Item Loading How likely would you tell Rachel that it would be better for her children if they did not see her in prison? How likely would you encourage Rachel not to have any contact with her childrne while in . rison? 80 muslin Scenario Five measured a respondent's behavioral intention to interact with inmates in an emotional situation (see Appendix 0). A factor analysis resulted in two factors. Subsequent reliability analyses resulted with one factor having a low coefficient alpha (i.e., .560). Therefore this factor and the corresponding test items were eliminated from further analysis. The remaining factor consisted of four items: "Scenario Five." The coefficient alpha was .875 (see Table 15). A low score indicated a respondent's intention to become involved in an emotional situation. Table 1 5 Scenario Five (N=138) . Factor Test Item ‘ Loading How likely would you ask Shannon what was wrong? How likely would you attempt to comfort Shannon? How likely would you suggest that Shannon talk to someone? How likel would ou sit down and talk to Shannon? ScenarisLSIx Scenario Six assessed a respondent's behavioral intention to treat a female inmate as a child (see Appendix P). The first factor consisted of four 81 items: "Scenario Six — Personal Involvement." One item was subsequently deleted to improve the coefficient alpha. The coefficient alpha was .859 (see Table 16). The scale ranged from 4 to 16 with a low score indicating that the respondent would be more likely to interact with the inmates in this situation. The second factor consisted of two items: "Scenario Six — Paternalistic Treatment." The coefficient alpha was .720 (see Table 17). The scale ranges from 2 to 8 with a higher score indicating that a respondent would be more likely to treat the inmates in a paternalistic manner. Table 16 Scenario Six — Personal Involvement (N=134) Factor Test Item Loading How likely would you ask Sandy and Karla what they were ar-ui 2 about? How likely would you try to calmly talk to Sandy and Karla? How likely would you take any formal action? How likel would ou attem -t to settle the are ument? 82 Table 1 7 Scenario Six - Paternalistic Treatment (N=141) Test Item How likely would you tell them that they were acting like children? How likely would you feel as though you have to handle the situation as thou-h ou were scoldin children? Table 18 is a summary of the scenario measures (i.e., behavioral intentions) and their corresponding reliabilities. Table 18 Summary of Scenario Measures and Coefficient Alpha Coefficient Alpha —-2- Scenario Two - Financial Independence Scenario Two - Work Inside the Home Scenario Two — Work Outside the Home Scenario Three Scenario Four — EncourageMaintainingTies Scenario Four — Discoura-e Maintainin- _ Scenario Six - Personal Involvement .859 Scenario Six - Paternalistic Treatment .720 83 Analysis of the Data The analyses for this research are presented in the next four chapters. Chapter Four provides a summary of each of the four groups on general background factors such as age, gender, racial/ethnic background. Specific information pertaining to each group is also summarized such as current offense serving (i.e., female inmates and peer counselors); program services offered to female inmates (i.e., program staff); and length of employment at Bedford Hills Correctional Facility (i.e., program staff and correctional officers). As mentioned previously in the literature review, studies have revealed a significant relation between sex, age, and attitudes toward women. Chapter Five examines the association of sex and age on a general attitude toward women as well as the specific attitudinal measures pertaining to female inmates. Chapter Six answers the following research question: "To what extent do stereotypes of female inmates differ by group membership, controlling for social distance?" In order to test the relation between group membership and attitudinal measures, analysis of variance was implemented, controlling for social distance. Chapter Seven answers the following research question: "'l'o what extent do specific attitudes toward female inmates and a general attitude toward women influence behavioral intentions?" Multiple regression analyses were implemented in order to examine how much of the behavioral intentions' 84 variance can be explained by specific attitudinal measures (i.e., female inmates) and how much of the variance can be explained by the global measure (i.e., women). 85 ENDNOTES Data collection started on 25 April 1996. The correctional officer group was retained in this study for subsequent analyses. As stated previously, more officers refused to participate in the study rather than complete the questionnaire. The researcher did not probe either the participants or the non-participants as to the reasoning for their decision. However, it is essential to emphasize a potential caveat bias with this group. One coefficient of reproducibility (i.e., .95) falls within the acceptable range while the second score (i.e., .87) is slightly less. The difference between these two coefficients is how the mixed types are scored. Thus, this resulted in different methods for calculating the number of errors for the scale. For all of the factor analyses, the method of factor extraction was principal components. Instead of implementing the default criterion of including only those factors with an eigenvalue of 1.0 or greater, 3 scree test was used to assess the number of factors in the analyses. An oblique rotation was used for the analyses. This method was implemented because the factors are likely to be correlated (Norusis, 1985). From the factor analyses, various scales were constructed based on variables with high factor loadings (Kim 8. Mueller, 1978). DeVelIis (1991) illustrated how to use factor analysis to develop scales. After the extraction and rotation of the factors, he examined the factor loadings. Each factor was defined by those items that load the highest. After examining those items, he attempted to distinguish the latent variable that was represented by each factor (p. 104). For the present study, this same procedure was implemented to obtain the attitudinal and behavioral measures. CHAPTER FOUR GENERAL CHARACTERIZATION OF THE FOUR GROUPS OF STUDY The following is a general description of the four groups that were surveyed: a) female inmates; b) peer counselors; c) correctional officers; and d) program staff. Descriptive factors include age, racial/ethnic background, gender as well as particular relevant variables for each group such as offender history (e.g., female inmates) and length of employment (e.g., correctional officers). At the end of this chapter, Table 19 compares and summarizes the general characteristics of these four groups. Female Inmates Within the group of female inmates, 74 responded to the questionnaire. The respondents ranged in age from 16 years to 59 years. The mean was 34.7 years of age. The largest age group was from 30 to 39 years of age (41.7%) followed by those between 21 to 29 years of age (25.0%) (see Figure 2). Comparing these distributions to the general prison population, the average age for those incarcerated in Bedford Hills was 33 years. Thirty seven percent of the women were between the ages of 30 to 39 with the next largest age group between the ages of 21 to 29 (34%) (Bedford Hills Correctional Facility, 1994). The 1991 Survey of Inmates of 86 87 State Correctional Facilities reported that 50% of the female inmates were between the ages of 25 to 34 years. The sample for the present study was slightly older than the national average. One reason may be due to Bedford Hills being a maximum security facility. Therefore, there are more serious offenders incarcerated for lengthier sentences (BJS, 1994). The largest racial/ethnic group within this sample was African American (36.1%) followed by Hispanic (30.6%) and white (22.2%) (see Figure 3). The racial/ethnic background of the participants in this study is similar to that reported by Bedford Hills (1994). The largest racial/ethnic group within the institution was African American (53%) followed by Hispanic (28%) and white (18%). ChildanLLEemaleJnmatas Fifty five of the 74 (74.3%) respondents reported having children. ME._.-r-- *—‘ The average number of children for each inmate was 2.7. The average age .___ of the children was 13.3 years old with a range from newborn to over 41 years. Over 75% of the children were 18 years or younger (see Figure 4). why Of those children 18 years or younger, slightly over 62% were livingflathome with their mothers prior to incarceration (see Figure 5). The national profile revealed that over three-quarters of all female inmates had children. Of these .._... women, two-thirds had children under the age of 18 years. Seventy-five 88 Figure 2 Age Distribution of Female Inmates 20 to 29 25.0% 7_Under 20 30 to 39 4.2% 41.7% 50 and Older 8.3% 40 to 49 20.8% Figure 3 Racial/Ethnic Background of Female Inmates African American 36.1% Hispanic Other 3.3% \Native American 2.8% N = 72 89 Figure 4 Age Dlatrlbutlon of Children Under 7 7 to 12 3. < g 13 to 18 >- 19 to 24 25 and Older 50 Number of Children N = 1 47 Figure 5 Children Living At Home Prior to Mothers' Incarceration (1 8 Years or Younger) Living At Home 62.5% ‘No Response 8.9% Not Living At Home N=112 28.6% 90 percent of these children were living with mothers prior to incarceration (BJS, 1994). El . IL)! I' IIE l IE I 1 Over 52% of the women surveyed had not completed high school or received a G.E.D. Slightly less than seven percent completed high school and an additional seven percent received a G.E.D. Thirteen of the inmates had some college education (18.0%) while an additional 8.1% received a bachelor's degree (see Figure 6). This sample of female inmates had a much lower educational level compared to the national profile which reported 58% of the female inmates had a high school diploma, G.E.D., or some college (BJS, 1994). Forty two (58.3%) of the respondents had some form of vocational programming. Slightly over 58% had received this training prior to being involved in the criminal justice system. Over 25% received vocational training while incarcerated. Over 22% had received training in clerical services followed by cosmetology (20.0%) (see Figure 7). There were some respondents who had participated in more non-traditional types of vocational programming such as printing (6.7%), machine operations (4.4%), and welding (2.2%). Slightly over 34% of the respondents reported that prior to incarceration their primary source of income was employment. Of these, Up to 8th Grade 9th to 11th Grade High School Graduate G.E.D. One to Three Years of College Bachelors Degree Five Years or More Other N=72 91 Figure 6 Educational Background 1 5 20 25 Number of Female Inmates Figure 7 Types of Vocational Programming Building Maintenance Business Clerical . Computers Cosmetology Machine Operators Nursing Assistants Printing Welding Other 17 w I I I I I I l I I l I I l I I . I l l I I 4 6 8 Number of Female Inmates 10 12 92 71.4% were employed on a full time basis. An additional 30.6% reported that their primary source of income was state assistance. Approximately 15% of the respondents' primary source of income was from family while an additional 1 1.1% received their income through a combination of the above sources (see Figure 8). The national profile of female inmates reported that 47% of the women were employed prior to incarceration which is higher than this study's sample (BJS, 1994). QflendeLtliatht Sixty three of the respondents listed the various offensels) which they were currently serving at the facility. There were a total of 81 offenses listed. Thirty seven percent of these offenses were drug related. Slightly less than 20% of the women were serving sentences for murder or manslaughter followed by robbery (12.3%) (see Figure 9). The type of offense is similar to the national profile. Almost 1 in 3 female inmates were incarcerated for a drug offense. Approximately 3 in 10 women were incarcerated for a violent offense (BJS, 1994). The respondents were also asked to list the length of sentencels) they received for these offenses. The sentence lengths were primarily listed with a minimum and maximum length of time. Slightly less than 60% received a minimum sentence of five years or less for the respective offense followed 93 Figure 8 Primary Source of Income Job 34.7% State Assistance 30.6% Other 8.3% N = 72 Combination I , 11.1% Farnily 15.3% Figure 9 Current Offense Arson Assault Burglary Conspiracy Criminal Sexual Conduct Drug Related Offense Larceny Murder/Manslaughter Robbery/Armed Robbery Weapons Other Number of Offenses N = 81 Offenses 35 94 by 20.7% of the offenses with a minimum sentence of six to ten years (see Figure 10). Lastly, the respondents were asked to list the month and year they were admitted to Bedford Hills. The dates of admission ranged from two months to almost thirteen years with an average of 2.83 years. Peer Counselors There were a total of 29 peer counselors surveyed for this study. The respondents ranged in age from 22 to 58 years. The mean was 36.7 years (see Figure 11). African Americans (33.3%) and Hispanics (33.3%) were the largest racial/ethnic group within this sample followed by whites (22.2%) (see Figure 12). thldanzLEennleJnmates Twenty two of the peer counselors surveyed reported having children (75.9%). The average number of children for each inmate was 2.0. The average age of the children was 17.6 years old with a range from 1 1 months to slightly less than 38 years. Over 50% of the children were 18 years or younger (see Figure 13). Of those children 18 years or younger, 81% were living at home with their mothers prior to incarceration. Three of the 21 children were not with their mothers prior to incarceration and one mother had her infant child living with her in the correctional facility (see Figure 14). 95 Figure 10 Minimum Sentence Serving 5 or Less 59.8% \ . . 26 or More 2.4% 6 to 10 20.7% 21 to 25' 4.9% 11 to 1.33.0 20 93% 2.4% N = 82 Sentences 96 Figure 1 1 Age Distribution of Peer Counselors 22 to 29 25.9% 30 to 39 37.0% \ 50 and Older 3.7% N = 27 \ 40 to 45 33.3% Figure 1 2 Racial/Ethnic Background of Peer Counselors African American 33.3% Hispanic Other 11.1% N = 27 l White 22.2% Years of Age 25 and Older N=41 97 Figure 1 3 Age Dlstrlbution of Children Under 7 I l l l l l 7to12 l l | l | l l l l l l 13to 18 191024 0 2 4 6 8 1 0 12 14 16 Number of Children Figure 1 4 Children lelng At Home Prlor to Mothers' Incarceration (18 Years or Younger) N=21 Living At Home 81.0% Living in Prison 4.8% \Not Living At Home 14.3% 98 El I' I'll l' “E I IE I I One of the criteria for working as a peer counselor is to have at least a high school diploma or a G.E.D. Three of the 29 peer counselors had at least a G.E.D. The remaining 26 peer counselors had some college education with almost 38% having a bachelors degree (see Figure 15). Twenty three (79.3%) of the peer counselors had some form of vocational programming. Ten (34.5%) of the respondents received this training while incarcerated while an additional ten had received this training prior to being involved in the criminal justice system (i.e., on the "outside"). Over 26% of the peer counselors had received some vocational training in business followed by clerical training (23.5%). Four (1 1.8%) of the respondents had some training in cosmetology as well as computers (see Figure 16). Slightly more than 34% of the respondents reported that prior to incarceration, their primary source of income was through employment. Of these, 55.6% were employed on a full time basis and 33.3% were employed part-time. Approximately 21% reported that their primary source of income was state assistance while an additional 31% reported their primary source of income was a combination of various sources such as job, family, and state assistance (see Figure 17). 99 Figure 1 5 Educational Background Two Years of College 207% One Year of College ‘ 10.3% G.E.D. Three Years of College 10.3% 10.3% Five Years or More 10.3% Bachelors Degree 37.9% Figure 1 6 Types of Vocational Programming Business Clerical Computers Cosmetology Electronics Legal Research Nurses Assistant Printing Other O 2 4 6 Number of Female inmates 100 Figure 17 Primary Source of Income State Assistance 20.7% Other 3.4% Family 1 0.3% N = 29 Combination 31.0% Figure 1 8 Current Offense Drug Related 28.6% Burglary 7% Other 1 1.4% Murder/Manslaughter 45.7% \ Robbery 8.6% N = 39 Offenses 101 QflendeLtlistm Twenty six of the respondents listed the various offense(s) which they were currently serving at the facility. There were a total of 39 offenses listed. Over 45% of the convictions were for murder or manslaughter offense. Over 28% of the offenses were drug related (see Figure 18). The respondents were also asked to list the length of sentence(s) they received for these offenses. The sentences were typically listed with a minimum and maximum length of time. Slightly more than 22% had received a minimum sentence of five years or less for their respective offense. Almost 28% of the offenses had a respective minimum sentence of six to ten years (see Figure 19). The respondents were also asked to list the month and year they were admitted to the correctional facility. The dates of admission ranged from five months to sixteen years with an average of 7.12 years. E E l' B 'l 'l'l' Forty percent of the peer counselors provided some type of tutorial services to female inmates. An additional 25% of the peer counselors provided vocational training services (see Figure 20). Twenty two of the 29 respondents reported how long they had been peer counselors. The length of time ranged from 2 months to thirteen years with an average of 4.4 years. 102 Figure 1 9 Minimum Sentence Serving 6to 10 27.8% 5 or Less 22.2% 21 or More 11 to 15 22.2% 19.4% 16 to 20 N = 36 Sentences 83% Figure 20 Peer Counseling Responsibilities Grievances 5.0% Tutoring 40.0% Family Education 1 5.0% More than One 10.0% Workshops 5.0% N = 2° Vocational Training 25.0% 103 C . I II I. IE [1 Compared to the national profile, this sample of peer counselors is older. However, they are similar in racial/ethnic background. While this sample is similar to those inmates in the national survey regarding children, these peer counselors had fewer children under the age of 18 years. This is probably due to these individuals being approximately three years older than those inmates in the national study. Since peer counselors are required to have at least a high school diploma or G.E.D., their educational level is higher. However, this sample reported a lower number of inmates being employed prior to incarceration. The greatest discrepancy between this sample and the national profile is offense. More peer counselors were incarcerated for a violent offense (i.e., murder or manslaughter) and fewer were serving for a drug offense. Correctional Officers Within the group of correctional officers, twenty one responded to the survey. The respondents' ages ranged from 22 to 62 years with a mean of 37.9 years (see Figure 21 ). The largest racial/ethnic group within this sample was white (57.1%) followed by African American (23.8%) and Hispanic (14.3%) (see Figure 22). Among the respondents, there were twelve females (57.1%) and nine males (42.9%). This breakdown of sex is 104 Figure 21 Age Distribution of Correctional Officers 22 to 29 / 33.3% 30t039 19.0% 50 and Older 1 4.3% N = 2‘ 4010 49 33.3% Figure 22 Racial/Ethnic Background of Correctional Officers Hispanic 14.3% African American 23.8% “Other 4.8% 57.1% 105 similar to the entire group of correctional officers at Bedford Hills. There are 190 (56.4%) males and 147 (43.6%) females. Approximately 81% of the respondents had at least one year of college with two officers having a bachelors degree (see Figure 23). EmnloxmanLBackgmund Respondents were employed as correctional officers an average of 9.7 years with a range of 9 months to 28 years. The average number of years the respondents were employed as correctional officers at Bedford Hills Correctional Facility was 7.1 years. The average number of years for the entire population of correctional officers is six years. Of the 21 respondents, twelve (57.1%) were employed at some time in a male facility. Among the twelve respondents, the average number of years employed in a male facility was 3.8 years with a range from 2 months to 15 years. As mentioned in Chapter Three, it was difficult to obtain an equal representation of correctional officers from the various shifts (i.e., morning, afternoon, and evening). Therefore, the largest number of respondents worked the morning shift (66.7%) followed by the afternoon shift (23.8%) (see Figure 24). Program Staff Within the group of program staff, 27 responded to the survey. The respondents ranged in age from 34 years to 64 years. The mean was 46.9 106 Figure 23 Educational Background One Year of College 14.3% High School Graduate 1 9.0% Two Years of College Five Years Or More 38.1 % 4.8% Bachelors Degree 4.8% Three Years of College N — 21 19.0% Figure 24 Correctional Officers Work Shift Morning 66.7% More than One 4.8% \Evening 4.8% N = 21 Afternoon 23.8% 107 years of age. Half of the program staff were between the ages of 40 to 49 years (see Figure 25). The largest racial/ethnic group within this sample was white (69.2%) with the next largest group being Hispanic (15.4%) (see Figure 26). Among the respondents, there were fifteen females (55.6%) and twelve males (44.4%). Sixty three percent of the respondents had five years or more of college while an additional 14.8% had a bachelors degree (see Figure 27). EmolnxmenLEacknmund Respondents were employed as program staff an average of 8.4 years with a range of 5 months to 22 years. The average number of years the respondents were employed as program staff at Bedford Hills Correctional Facility was 7.3 years. Of the 27 respondents, eight were employed at some time in either an all male correctional facility (N = 7) or a co-educational facility (N =1) (see Figure 28). Among these eight respondents, the average number of years employed in a male facility was 6.5 years with a range from 4 months to ten years. The various services provided by these program staff are illustrated in Figure 29. The majority of respondents were in the counseling unit at Bedford Hills (34.6%) followed by those in the educational unit (26.9%) and vocational unit (11.5%). Those respondents included in the "Other" category provided various services such as substance and alcohol abuse 108 Figure 25 Age Distribution of Program Staff 60 and Older 7.7% NT” mow 26.9% Fig u re 26 Racial/Ethnic Background of Program Staff Hispanic African American 7.7% Other 3.8% Native American 3.8% N = 26 White 69.2% 109 program, recreational program, grievance program, religious services, library services, assessment program, and mental health. 110 Figure 27 Educational Background Bachelors Degree 1 4.8% 3 Years of College 3.7% 2 Years of College 11.1% 5 Years or More 63.0% N=27 Figure 28 Employed at a Male Facility 111 Figure 29 Services Provided To lnmates Counseling 34.6% N ' 27 Educational Vocational 26.9% 11.5% 112 Table 1 9 Summary and Comparison of Background Information Female Peer Correctional Program Inmates Counselors Officers Staff (N = 74) (N = 29) (N = 21) (N = 27) Mean Age 34.7 36.7 37.9 46.9 Racial/Ethnic Background African American 36.1% 33.3% 23.8% 7.7% Hispanic 30.6% 33.3% 14.3% 15.4% White 22.2% 22.2% 57.1% 69.2% Native American 2.8% 3.8% Other 8.3% 11.1% 4.8% 3.8% Children Have Children 74.3% 75.9% Average number 2.7 2.0 Average age 1 3.3 1 7.6 Living with prior to incarceration 62.5% 81 .0% Education Level Up to 8th grade 9.7% 9th to 11th 43.0% H.S. graduate 6.9% 14.3% G.E.D. 6.9% 10.3% 4.8% 1 to 3 yrs college 18.0% 41.3% 71.4% 14.8% Bachelors degree 8.3% 37.9% 4.8% 14.8% Five yrs or more 4.2% 10.3% 4.8% 63.0% Other 2.8% 7.4% Vocational Training 58.3% 79.3% 113 Table 19 (continued) Female Peer Correctional Program Inmates Counselors Officers Staff lN=74l (N=29l (N=21) (N=27) Primary Source of Income Job 34.7% 34.5% State Assistance 30.6% 20.7% Family 15.3% 10.3% Combination 11.1 % 31.0% Other 8.3% 3.4% Average Time Served 2.8 years 7.1 years Gender Female 57.1 % 55.6% Male 42.9% 44.4% Average Number of Years in Profession 9.7 8.4 Average Number of Years at Bedford Hills 7.1 7.3 Previously Employed at Male Facility 57.1% 25.9% CHAPTER FIVE GENERAL FACTORS INFLUENCING ATTITUDINAL MEASURES This chapter focuses on general factors that may be associated with a respondent's attitudes. First, the relationship of sex, age, and educational level to the various attitude measures are examined: a) Sexist Attitudes Toward Women Scale; b) Attitudes about Non-Traditional Work for Female Inmates; c) Female Inmates' Need to Work; d) Maintaining Ties with Children; e) ldealizations of Motherhood; f) Child Visitation; 9) Paternalistic Attitudes Toward Female Inmates; and h) Female Inmates Are irrational. Second, as mentioned in Chapter Three, to assess if there were attitudinal differences toward male and female inmates, scales were constructed in order to make such comparisons. These attitude comparisons are also examined in this chapter.‘ Sexist Attitudes Toward Women A one-way analysis of variance tested if the sex of a respondent was related to his or her attitudes towards women. The results revealed that there was no significant difference between male and female respondents and their Sexist Attitudes Towards Women Scale scores (F = .145, r]2 = .001 , p = .704). There was no significant correlation between an individual's age 114 115 and his or her Sexist Attitudes Towards Women Scale scores (r =-.018, p = .845). There was a significant, indirect correlation between an individual’s educational level and his or her Sexist Attitudes Toward Women Scale scores (r=-.340, p = .000). The lower an individual’s educational level, the more likely she or he had sexist attitudes toward women. Attitudes About Non-Traditional Work For Female Inmates A one way analysis of variance for sex revealed no significant difference between males and females and their Attitudes About Non- Traditiana/ Work For Female Inmates (F = .548, r12 = .005, p = .461 ). There was no significant correlation between age as well (r= .035, p = .696). Again, educational level was significantly correlated with Attitudes About Non-Traditional Work For Female Inmates (r= .306, p = .000). The higher an individual's educational level, the more likely she or he would encourage inmates to pursue non-traditional occupations. Female Inmates' Need to Work A one-way analysis of variance for sex revealed a significant difference between males and females (F=4.81, n2: .040, p= .030). Female respondents were more likely to have the attitude that women prisoners do not need to be financially independent compared to male respondents (see Figure 30). There was a significant inverse correlation between age as well 116 (r=-.23, p = .006). The older an individual, the more likely she or he had a lower score on the Female Inmates' Need to Work Scale (i.e., the more likely they perceived that female inmates do need to be financially independent). Educational level was also significantly related (r=-.449, p = .001). The higher the educational level, the more likely a person perceived a female inmate’s need to be financially independent. Maintaining Ties With Children A one-way analysis of variance for sex revealed a significant difference between male and female respondents and their scores on the Maintaining Ties With Children Scale (F =4.75, n2: .039, p = .031 ). Females were more likely to have the attitude that mothers should maintain ties with their children compared to males (see Figure 31). There was no significant correlation between age and attitudes about female inmates maintaining ties with their children (r= .145, p= .085). However, there was a significant, direct correlation with educational level and Maintaining Ties With Children Scale (r= .221 , p = .008). In this instance, the higher educational level, the more likely a person’s attitude] that a child should not maintain ties with his or her mother. in prison. 117 Figure 30 Sex and Attitudes About Female Inmates' Need to Work Females 2‘3 :13 Males i 0 2 4 6 a 10 12 Females (N—95) Males (N—21) Mean Score Figure 31 Sex and Attitudes About Maintaining Ties With Children Females E 2 i a: ‘5 )1 r3 Males i I 0 2 4 6 8 10 12 14 118 ldealizations of Motherhood A one-way analysis of variance for sex resulted in a significant difference between male and female respondents (F: 13.5, r]2 = .108, p = .0004). Male respondents were more likely to have negative attitudes about mothers in prison compared to female respondents (see Figure 32). There was a significant direct correlation with age (r=.221, p =.013). The older an individual, the more likely she or he had more negative attitudes toward female inmates and motherhood. Educational level was not significantly correlated (r=.163, p = .055). Child Visitation Scale When comparing the mean scores between male and female respondents, there was a significant difference (F: 5.68, n2 = .047, p =.019). Male respondents were more likely to have negative attitudes about children visiting their mothers in prison compared to female respondents (see Figure 33). There was no significant correlation with age (r=.122, p-.151) or educational level (r=.065, p= .440). Paternalistic Attitudes Toward Female Inmates There was no significant relation between sex and Paternalistic Attitudes Toward Female Inmates (F=3.02, n2: .027, p= .085). There was 119 Figure 32 Sex and Attitudes About ldealizations of Motherhood Females Sex of Respondent I 0 5 10 15 Females (N=95) Males (N=19) Mean Score Figure 33 Sex and Attitudes About Child Visitation i i l | l 1 1 I Females Sex of Respondent 0 1 2 3 4 5 Females (N—ec) Males (N=21) Mean Score 120 a significant direct correlation with age (r= .278, p= .001). The older an individual, the more likely she or he would have paternalistic attitudes toward female inmates. There was no significant correlation with educational level (r = .079, p = .353). Female Inmates Are Irrational There was a significant difference between male and female respondents and their attitudes about female inmates being irrational (F=3.91, r12: .034, p= .050). Male respondents were more likely to have stronger attitudes that female inmates are irrational when compared to female respondents (see Figure 34). There was no significant correlation with age (r=.030, p=.731) or educational level (r=.052, p=.539). Figure 34 Sex and Attitudes That Female Inmates Are Irrational Females Sex of Respondent Males I I o 2 4 6 8 1o 12 14 Females (N=94) Males (51-21) Mean Score 121 Summary Tables 20, 21, and 22 summarize the effects of respondents' sex , age, and educational level on the various attitudinal measures. Sex did not have a significant effect on the global attitude measure (i.e., Sexist Attitudes Toward Women Scale). The respondents' sex did have a significant effect on five of the seven specific measures concerning attitudes toward female inmates. Of the five measures, the male respondents had more negative attitudes than the female respondents on four of these scales. However, compared to male respondents, female respondents were less likely to perceive female inmates' need to work. The respondents' age was not significantly correlated with their Sexist Attitudes Toward Women Scale. Age was significantly correlated with three of the seven specific attitude measures (i.e., Female Inmates Need to Work, ldealiza tions of Motherhood, and Pa ternalis tic A ttitudes Toward Female Inmates). Of the three scales, the older the respondent, the more negative his or her attitudes towards female inmates on two of these scales. However, the younger the respondent, the less likely they perceive a female inmates' need to work. 1 22 Table 20 Summary of the Effects of Sex On Attitudinal Measures I Scales F n2 Significance j Sexist Attitudes Toward Women Scale (N = 104) 0.15 .001 .704 Attitudes About Non-Traditional Work for Female Inmates (N =106) 0.55 .005 .461 Female Inmates' Need to Work (N =115) 4.81 .040 .030 Maintaining Ties With Children (N =117) 4.75 .039 .031 ldealizations of Motherhood (N =1 13) 13.5 .108 .000 Child Visitation (N = 1 16) 5.68 .047 .019 Paternalistic Attitudes Toward Female inmates (N = 1 12) 3.02 .027 .085 Female inmates Are Irrational (N =1 14) 3.91 .034 .050 Educational level was significantly correlated with the general attitudinal measure, Sexist Attitudes Toward Women Scale, and three of the seven specific attitudinal measures (i.e., Attitudes About Non-Traditional Work for Female Inmates, Female lnma tes ’ Need to Work, and Maintaining Ties with Children). in three of these four measures, the higher an individual's educational level, the more likely she or he had less stereotypical 123 Table 21 Summary of the Effects of Age On Attitudinal Measures s..- r Sexist Attitudes Toward r I Significance Women Scale (N = 124) -.018 .845 Attitudes About Non-Traditional Work for Female inmates (N =129) .035 .696 Female inmates' Need to Work (N = 140) -.230 .006 Maintaining Ties With Children (N = 142) .145 .085 ldealizations of Motherhood (N =137) .221 .013 Child Visitation (N =141) .122 .151 Paternalistic Attitudes Toward Female inmates (N =136) .278 .001 Female Inmates Are irrational (N = 139) .030 .731 attitudes about female inmates. However, individual’s with a higher educational level were less likely to have attitudes encouraging children to maintain ties with their mothers in prison. 1 24 Table 22 Summary of the Effects of Educational Level on Attitudinal Measures Scales l r I Significance I Sexist Attitudes Toward Women Scale (N =124) -.340 .000 I Attitudes About Non-Traditional Work for Female inmates (N =132) .306 .000 Female inmates' Need to Work (N =143) -.449 .000 Maintaining Ties With Children (N =144) .221 .008 ldealizations of Motherhood (N = 140) .163 .055 Child Visitation (N = 144) .065 .440 Paternalistic Attitudes Toward Female inmates (N = 140) .079 .353 Female inmates Are Irrational (N =142) .052 .539 Attitudes Toward Male and Female Inmates As mentioned in Chapter Three, corresponding specific attitude measures for male inmates were also constructed based on the scales pertaining to female inmates. These scales were developed in order to determine if there were differences pertaining to the sex of the inmate. 125 There was a significant difference when comparing whether a female inmate, or a male inmate, should maintaining ties with her or his children (t =-4.39, p = .000). Respondents perceived that it was more important for female inmates to maintain ties with children compared to male inmates maintaining ties with their children. There were significant differences when comparing ldealizations of parenthood between female inmates and male inmates (t=-3.37, p =.001). Respondents had more negative parenthood ldealizations about fathers in prison when compared to mothers in prison. Respondents had significant differences concerning children visiting their mothers in prison or their fathers in prison (t=-2.47, p =.015). Children visiting their fathers in prison was perceived more negatively than children visiting their mothers in prison. Finally, when comparing paternalistic attitudes toward inmates, there were significant differences (t=4.58, p = .000). Respondents had more paternalistic attitudes toward female inmates than male inmates. Table 23 summarizes six of the seven specific attitudinal scales2 by examining if the sex of the inmate was a significant factor among respondents. The sex of the inmate was a significant factor on four of the six measures. Three of the four measures refer to parents in prison. There were more positive attitudes concerning mothers in prison compared to attitudes about fathers in prison. Alternatively, the sex of the inmate was not a significant factor concerning a respondent's perception of an inmates' 1 26 Table 23 Comparing Specific Attitudinal Measures Between Female Inmates and Male inmates Scale l Mean t Value Significance. Female Inmates' Need to Work ' (N = 130) Female Inmates 9.65 -1.01 .313 Male Inmates 9.85 Maintaining Ties With Children (N = 1 28) Female inmates Male inmates ldealizations of Motherhood (N = 1 27) Female Inmates 14.42 -3.37 .001 Male inmates ) Child Visitation ' (N=132) Female Inmates 3.95 Male inmates J Paternalistic Attitudes Towards Female Inmates (N=122) 16.43 4.58 .000 i Female inmates 14.67 l Male Inmates i Female Inmates Are Irrational (N =124) Female Inmates 1 1.60 -1.39 .169 Male Inmates 1 1.95 E 127 need to work or a respondent's perception that inmates are irrational.Three of the four measures refer to parents in prison. There were more positive attitudes concerning mothers in prison compared to attitudes about fathers in prison. Alternatively, the sex of the inmate was not a significant factor concerning a respondent's perception of an inmates' need to work or a respondent's perception that inmates are irrational. 1 28 ENDNOTES This sample was not selected to study the effects of age and gender simultaneously with social distance and group membership (i.e., Chapter Six). First, there was no sex variation in two of the four groups (i.e., female inmates and peer counselors). Second, the sample was not stratified to include various age groups. initially, a four-way ANOVA revealed that interactions of the four factors were not possible due to empty cells. The only results were the main effects of these four variables on the dependent variables. Therefore, the relationship of age and sex to the various attitude measures are examined in this Chapter. The effects of sex (i.e., among correctional officers and program staff) is examined further in Chapter Seven (i.e., the behavioral intention measures). There was no corresponding scale constructed pertaining to non- traditional work for male inmates. CHAPTER SIX GROUP MEMBERSHIP AND STEREOTYPES OF FEMALE INMATES As mentioned previously, this chapter answers the following general research question: "To what extent do stereotypes of female inmates differ by group membership, controlling for social distance." Prior to addressing this general research question, the initial analysis focused on whether there was a significant difference between social distance and group membership. The analysis revealed that these two factors were significantly related. Thus, for the remaining analyses, it was essential to partition the effects of social distance and group membership on the various attitudinal measures. Specifically, social distance and group membership operated as independent variables on the attitudinal measures. Group Membership and Social Distance -- Female Inmates Although the Social Distance Scale measures interactions between the various groups that were surveyed, the only portion which is the focus of the present study are those interactions with the group of female inmates. The scores on the Social Distance Scale -- Female Inmates ranged from 0 to 6 with a mean of 3.9. Again, a higher score indicates closer interactions with female inmates. The group with the highest mean score was peer counselors 129 130 (2 =4.89) followed by program staff (52 =4.31) and female inmates (2 =3.52). The group with the lowest mean score was correctional officers (52 =3.25) (see Figure 35). A one-way analysis of variance revealed that these mean scores among the groups were significant (F=4.66, r12 = .094, p = .004) (see Table 24). it may initially appear problematic as to why female inmates have a relatively greater social distance with other female inmates compared to some of the other groups. However, Poliock-Byrne (1990) noted that ”(female inmates] feel isolated and surrounded by uncaring or hostile others, and the loss of their children is cause for acute worry and depression (p. 129)." Thus, it is essential to contextualize social distance among female inmates within an institutional setting in which establishing and maintaining supportive relationships is difficult. Sexist Attitudes Towards Women For this portion of the analysis, the following research question was examined: is group membership related to attitudes toward women? Again, The Sexist Attitudes Toward Women Scale ranges from 40 to 280 with a higher score indicating more sexist attitudes towards women. The sample scores ranged from 41 to 206 with an average of 1 12. Female inmates Peer Counselors Correctional Officers Group Membership Program Staff Mean Social Distance Scores 131 Figure 35 N = 138 Mean Score Table 24 Social Distance Among Group Membership (N = 138) Mean Source SS df Square F r Between groups 48.94 3 16.31 4.66 * .306 Within groups 472.44 135 3.5 Total 521.38 138 *p<.01 132 A simple factorial ANOVA model was used to test this research question‘ (Bohrnstedt & Knoke, 1988; Lomax, 1992; Norusis, 1990). Social distance had a significant linear effect on Sexist Attitudes Toward Women (F = 24.79, p = .000). Group membership also had a significant main effect (F =3.37, p= .021 ). This model explained 23.5% of the variance in Sexist Attitudes Toward Women Scale (see Table 25). Figure 36 illustrates the mean scores for each of the four groups. The lowest score, or least sexist attitudes towards women, was among the program staff (52 =92.5) followed by peer counselors (>‘< =101 .5). The highest mean score, or greater sexist attitudes, was among the female inmates (x = 1 18.6) followed by correctional officers (x =1 16.5). Attitudes about Non-Traditional Work for Female Inmates The next analysis addressed the following research question: is group membership related to attitudes about traditional and non-traditional work for female inmates? The Attitudes About Non- Traditional Work for Female Inmates scores ranged from 22 to 44 with a high score indicating encouragement of inmates to pursue various occupations regardless of gender. The sample scores ranged from 22 to 44 with a mean of 40.3. Group Membership 133 Figure 36 Mean Sexist Attitudes Towards Women Scores Female inmates Peer Counselors Correctional Officers Program Staff 0 20 40 60 80 100 120 140 N - 1 18 Mean Score Table 25 Sexist Attitudes Towards Women By Group Membership and Social Distance (N = 1 17) Source of Mean Variation SS df Square F r Covariate Social Distance 17679.49 1 17670.49 24.79 * Main Effects Group Membership 7191.10 3 2399.04 3.37** Total 105432.48 117 901.13 .486 * p<.000; ** p<.05 134 Social distance had a significant linear effect (F = 1 2.36, p = .001) (see Table 26). However, group membership did not have a significant main effect (F =1.94, p =.128). This model explained 13.1% of the variance of A ttitudes About Non- Traditional Work for Female Inmates. Table 26 Attitudes About Non-Traditional Work for Female Inmates By Group Membership and Social Distance (N = 124) Source of Mean Variation SS df Square F r Covariate Social Distance 399.46 1 399.46 12.36 * Main Effects Group Membership 187.55 3 62.52 1.94 Total 4465.07 124 36.01 .363 * p=.001 Female Inmates' Need to Work To examine one facet of vocational programming, the following research question was raised: is group membership related to attitudes about female inmates' need to work? The following analysis addressed this question. The Female Inmates Need to Work scale ranges from 3 to 15 with a higher score indicating a respondent's 135 attitude that female inmates do not need to be financially independent. The sample scores ranged from 3 to 15 with a mean score of 9.5. Table 27 is a breakdown of the individual test items and the corresponding mean for each group. Table 27 Mean Group Scores for Individual Test Items on the Female Inmates' Need to Work * Test Female Peer Correctional Program Item Inmates Counselors Officers Staff Many women in prison can find someone, such as friends or family, to help them financially when released. 3.58 2.66 3.29 2.78 Upon release, female inmates usually have the option of working outside the home or getting financially support so they can care for family members. 4.00 3.31 3.24 2.81 Women in prison will have a significant other in their life that will help them financially when released from prison. 3.07 2.38 2.57 2.19 * High score indicates attitude that female inmates do not need to be financially independent. Social distance did have a significant linear effect (F = 15.68, p = .000). Group membership also had a significant effect (F=9.65, p = .000). This model explained 25.7% of the variance of Female Inmates’ Need to Work (see Table 28). Female inmates were the group that were 136 Figure 37 Mean Female inmates’ Need To Work Scores Female Inmates .2- .C e Peer Counselors 3 E :1 o. 3 Correctional Officers (9 Program Staff 12 N = 1 34 Mean Score Table 28 Female Inmates’ Need to Work By Group Membership and Social Distance (N = 133) Source of Mean Variation SS df Square F r Covariate Social Distance 94.39 1 94.39 15.68 * Main Effects Group Membership 174.17 3 58.06 9.65 * Total 1044.90 133 7.86 .507 * p = 000 137 more likely to have the attitude that female inmates did not need to be financially independent compared to any other group (>‘< = 10.8) followed by correctional officers (>‘< =9.10). The group that was most likely to have the attitude that female inmates did need to be financially independent was program staff ()7 = 7.73) (see Figure 37). ldealizations of Motherhood For the next portion, the analysis addressed the following research question: is group membership related to attitudes about inmate mothers being "bad" mothers? The ldealizations of Motherhood scale ranges from 5 to 25 with a higher score indicating more negative attitudes about female inmates and motherhood. The sample scores ranged from 6 to 25 with a mean score of 14.4. Table 29 is a breakdown of the individual test items and the corresponding mean for each group. Social distance did not have a significant linear effect on ldealizations of Motherhood (F=3.18, p= .077). Group membership, however, did have a significant effect (F =7.1 8, p= .000). This model explained 16.2% of the variance of ldealizations of Motherhood (see Table 30). The groups that had the most negative attitudes about mothers in prison were correctional officers (2 = 16.7) and program staff (2 = 16.7). The group with the most 138 Table 29 Mean Group Scores for Individual Test items on ldealizations of Motherhood * Test Female Peer Correctional Program item inmates Counselors Officers Staff Many mothers in prison have a hard time being good parents to their children. 2.84 2.76 3.81 3.78 Even though a mother is in prison, it does not mean she is a bad parent. 1.45 1.18 2.20 2.19 People who are incarcerated tend to have some problems being good parents. 2.51 2.62 3.43 3.70 Most female inmates were not very good mothers prior to being locked up. 2.93 2.76 3.00 3.19 The vast majority of mothers in prison need to learn the basics of how to be an adequate parent. 3.66 3.59 4.05 3.85 ' High score indicates more negative attitudes about female inmates and motherhood. positive attitudes about mothers in prison was peer counselors (>7 = 13.0) and female inmates (>'< =13.5) (see Figure 38). Maintaining Ties With Children For the next two analyses, the following research question was tested: is group membership related to attitudes about maintaining ties between inmate mothers and their children? Group Membership 1 39 Table 30 ldealizations of Motherhood By Group Membership and Social Distance (N = 132) Source of Mean Variation SS df Square F r Covariate Social Distance 49.40 1 49.40 3.18 Main Effects Group Membership 334.75 3 111.58 7.18 * Total 2374.08 132 17.99 .402 * p =000 Figure 38 Female inmates Peer Counselors Correctional Officers Mean ldealizations of Motherhood Scores Program Staff N=133 20 Mean Score 140 The scores for Maintaining Ties With Children range from 5 to 25 with a higher score indicating that female inmates should not maintain ties with their children. The sample scores ranged from 5 to 25 with a mean score of 9.8. Table 31 is a breakdown of the individual test items for the Maintaining Ties with Children Scale and the corresponding mean for each group. Social distance did have a significant linear effect (F =6.53, p=.012). Group membership also had a significant main effect (F: 1 1.08, p = .000). This model explained 23.2% of the variance of Maintaining Ties With Children (see Table 32). The group that had the strongest attitudes about female inmates maintaining ties with their children was peer counselors (x = 7.9) followed by female inmates themselves (2 =8.9). Correctional officers had the weakest attitudes about female inmates maintaining ties with their children (>'< = 12.6) followed by program staff ()7 = 12.2) (see Figure 39). Child Visitation Scale The Child Visitation Scale ranges from 2 to 10 with a higher score indicating a more negative attitude about children visiting their mothers in 141 Table 31 Mean Group Scores for Individual Test items on Maintaining Ties With Children * Test Female Peer Correctional Program Item inmates Counselors Officers Staff it is always important for children to visit their mothers in prison. 1.82 1.29 2.48 2.56 Upon release, most female inmates will be good mothers to their children. 2.12 2.24 3.29 3.19 it is important for a mother to be reunited with her children upon release. 1.68 1.72 2.48 2.46 Maintaining contact between mothers in prison and their children helps them reunite upon release. 1.55 1.38 1.95 1.93 Children need to see their mothers when their mothers are in prison. 1.68 1.28 2.38 2.11 * High score indicates a respondent's attitude that female inmates should not maintaining ties with their children. prison. The sample scores ranged from 2 to 10 with a mean score of 3.9. Table 33 is a breakdown of the individual test items and the corresponding mean for each group. Social distance did have a significant linear effect (F = 7.70, p = .006). Group membership also had a significant main effect (F =4.34, p= .006). This model explained 13.7% of the variance of the Child Visitation Scale (see Group Membership and Social Distance 1 42 Table 32 Maintaining Ties With Children (N = 136) Source of Mean Variation SS df Square F r Covariate Social Distance 84.20 1 84.20 6.53 * Main Effects Group Membership 428.34 3 142.78 11.08 ** Total 2213.64 136 16.28 .481 * p<.05; ** p=.000 Figure 39 Mean Maintaining Ties With Children Scores Female inmates .9- fi Peer Counselors 1 2 o. a Correctional Officers (5 Program Staff N=137 Mean Score 143 see Table 34). The group that had the most negative attitudes concerning children visiting their mothers in prison was program staff (52 =4.7) followed by correctional officers (7 =4.6). The group with the most positive attitudes about children visiting their mothers in prison was peer counselors (x = 2.9) and female inmates (x =3.9) (see Figure 40). Table 33 Mean Group Scores for Individual Test Items on Child Visitation Scale’ Test Female Peer Correctional Program Item inmates Counselors Officers Staff Children should not be exposed to seeing their mothers in prison. 1.94 1.48 2.29 2.33 Children should have just limited contact with their mothers while they are incarcerated. 1.85 1.38 2.45 2.44 ’ Higher score indicates a more negative attitude about children visiting their mothers in pfison. Paternalistic Attitudes Toward Female Inmates This analysis examined the following research question: Is group membership related to attitudes about female inmates being child-like? 1 44 Table 34 Child Visitation Scale Group Membership and Social Distance (N = 134) Source of Mean Variation SS df Square F r Covariate Social Distance 27.66 1 27.66 7.70 * Main Effects Group Membership 46.76 3 15.59 4.34 * Total 541.73 134 4.04 .371 * p< .01 Figure 40 Mean Child Visitation Scores Female inmates .9- 5 Peer Counselors .8 § 2 a. a Correctional Officers 0 Program Staff 0 1 2 3 4 Mean Score N=135 145 The Paternalistic Attitudes Toward Female Inmates scores range from 5 to 25 with a higher score indicating more paternalistic attitudes towards female inmates. The sample scores ranged from 6 to 25 with a mean score of 16.3. Table 35 is a breakdown of the individual test items for Paternalistic A ttitudes Toward Female Inma tes. Table 35 Mean Group Scores for individual Test Items on Paternalistic Attitudes Toward Female Inmates * Test Female Peer Correctional Program item Inmates Counselors Officers Staff Women in prison are as responsible as other adults. 3.10 2.82 3.10 2.65 Women in prison are a lot like children. 2.50 2.17 3.29 3.33 Some women in prison throw ”temper tantrums." 3.72 3.72 4.33 4.19 Women in prison have a tendency to overreact. 3.11 2.83 4.00 3.41 Many women in prison complain too much. 3.21 3.21 4.30 4.0 * Higher score indicates a more paternalistic attitudes toward female inmates. Social distance had a significant linear effect on Paternalistic Attitudes Toward Female Inmates (F =9.25, p= .003). Group membership also had a significant main effect (F=4.79, p= .003). This model explained 15.9% of the variance of Paternalistic Attitudes Toward Female Inmates (see Table 146 36). The group that had the most paternalistic attitudes toward female inmates was correctional officers (>7 = 19.2) followed by program staff (2 = 17.4). The group with the weakest paternalistic attitudes toward female inmates was peer counselors (x = 15.2) and female inmates (x = 15.7) (see Figure 41 ). Female Inmates Are Irrational The last analysis for this chapter tested the following research question: is group membership related to attitudes about female inmates being irrational? The Female Inmates Are Irrational scores range from 4 to 20 with a higher score indicating a respondent's attitude that female inmates are irrational. The sample scores ranged from 4 to 20 with a mean score of 1 1.8. Table 37 is a breakdown of the individual items for the corresponding group. Social distance had a significant linear effect (F=6.02, p=.016). Group membership also had a significant main effect (F =3.99, p= .009). This model explained 12.2% of the variance of Female Inmates Are Irrational Scale (see Table 38). The group that had the strongest attitudes that female inmates are irrational was correctional officers (2 = 13.9) followed by 147 Table 36 Paternalistic Attitudes Toward Female inmates by Group Membership and Social Distance (N = 129) Source of Mean Variation SS df Square F r Covariate Social Distance 139.18 1 139.18 9.25 * Main Effects Group Membership 216.06 3 72.02 4.79 * Total 1880.16 129 17.33 .399 * p<.01 Figure 41 Mean Patemallstic Attitudes Toward Female Inmates Scores Female inmates .9- fi Peer Counselors 3 E i Q E Correctional Officers (9 Program Staff 0 5 10 1 5 20 Mean Score N=130 1 48 Table 37 Mean Group Scores for Individual Test Items on Female Inmates Are Irrational * Test Female Peer Correctional Program Item Inmates Counselors Officers Staff Women in prison are usually rational. 2.36 2.52 2.90 2.78 Women in prison can rationally express their anger. 2.58 2.62 3.43 3.00 In most instances, when a conflict occurs between two female inmates, it is easy to calmly resolve the situation. 3.26 3.00 4.00 3.85 Many women in prison can reasonably deal with disappointment. 2.88 2.76 3.43 3.15 * Higher score indicates a respondent's attitude that female inmates are irrational. Table 38 Female Inmates Are Irrational By Group Membership and Social Distance IN = 1331 Source of Mean Variation SS df Square F r Covariate Social Distance 54.16 1 54.16 6.02 * Main Effects Group Membership 107.76 3 35.92 3.99 * * Total 1323.32 133 9.95 .350 * p<.05; ** p<.01 149 Figure 42 Mean Female Inmatee Are Irrational Scores Female Inmates Peer Counselors Correctional Officers Group Membership Program Staff N-134 program staff (2 = 12.4). The group with the weakest attitudes that female inmates are irrational was peer counselors (>‘<= 1 1.0) and female inmates (>‘< =1 1.4) (see Figure 42). 150 Summary Table 39 is a summary of the effects of group membership and social distance on the seven specific attitudinal measures. Group membership had a significant effect on all but one of the specific attitudinal measures (i.e., group membership was not significant on Attitudes About Non-Traditional Work for Female Inmates). Female inmates had the most traditional attitudes concerning women prisoners' need to be financially independent. Generally, however, correctional officers had more negative attitudes concerning mothers prison as well as having more paternalistic attitudes towards female inmates. Social distance had a significant effect on all but one of the seven specific attitudinal measures (i.e., ldealizations of Motherhood). 1 51 Table 39 Summary of the Effects of Group Membership and Social Distance on Specific Attitudinal Measures I ) Scale F ratio Significance r Attitudes About Non-Traditional Work for Female Inmates Group Membership _ Social Distance Female Inmates' Need to Work ‘ Group Membership 9.65 .000 I Q__S_cial Distance __ 15.68 .000 ldealizations of Motherhood .402 Group Membership 7.18 .000 Maintaining Ties With Children Group Membership Social Distance 11.08 Social Distance 3.18 .077 .481 .000 Child Visitation .371 Group Membership 4.34 .006 Social Distance 7.70 .006 Paternalistic Attitudes Toward Female inmates .399 Group Membership 4.79 .003 Social Distance 9.25 .003 Female Inmates Are irrational .350 Group Membership 3.99 .009 Social Distance 6.02 .016 1 52 ENDNOTES A simple factorial model was implemented to examine the separate effects of group membership and social distance on the general and specific attitudinal measures. The model for each analysis consisted of one factor (i.e., group membership) and one covariate (i.e., social distance). Specifically, a hierarchical model was used which enters the covariate first then enters the factor. Each model examines a) the main effect of group membership; and b) the linear effect of social distance on the attitudinal measure. The linear effect of social distance is evaluated separately from the main effect of group membership (Norusis, 1990). CHAPTER SEVEN STEREOTYPES OF FEMALE INMATES AND BEHAVIORAL INTENTIONS For this portion of the analyses, the following general research question is addressed: "To what extent do specific attitudes toward female inmates and general attitudes toward women influence behavioral intentions?" To assess how much these attitudinal measures influence behavior intentions, various multiple regression analyses were conducted (see Table 32). For each of these analyses, the general attitudinal measure (i.e., Sexist Attitudes Toward Women Scale) was included in the regression. Additional analyses included regressing sex and group membership (i.e., correctional officer or program staff) to assess if these factors also influenced behavioral intentions. Scenario One For this scenario, the following research question was examined: Are attitudes about traditional and non-traditional work for female inmates related to behavioral intentions to encourage or discourage an inmate to participate in non-traditional programming? Furthermore, to determine if there is a relation between the specific and general attitudinal measure on behavioral intentions, the following research question was explored: 153 154 How much of the variation within the behavioral situation of encouraging an inmate to participate in non-traditional programming can be explained by attitudes about traditional and non-traditional work for women and how much of the variation can be explained by global attitudes toward women? Table 40 Summary of Regression Analyses SPECIFIC ATTITUDINAL BEHAVIORAL INTENTION MEASURE MEASURE Attitudes about Non-Traditional Work for Female Inmates Scenario One Female Inmates Need to Work Scenario Two - Financial Independence — Work inside the Home - Work Outside the Home ldealizations of Motherhood Scenario Three Maintaining Ties with Children Scenario Four Child Visitation Scale - Encourage Maintaining Ties - Discourage Maintainino Ties Female Inmates are Irrational Scenario Five Paternalistic Attitudes Towards Female Scenario Six Female Inmates - Personal Involvement f - Paternalistic Treatment In order to address these research questions, multiple regression analyses were conducted. Attitudes About Non- Traditional Work for Female Inmates was first entered into the regression. This attitudinal measure explained a significant portion of the variance in Scenario One (3:149, F=19.23, p=.000). When the global measure, Sexist Attitudes Toward Women Scale, was entered into the 155 regression, there was a significant increase in the amount of variance explained (r‘2 change=.112, F change=16.57, p=.0001). The total amount of variance in Scenario One explained by both attitudinal measures was 26.12%. Table 41 Scenario One: Encourage Participation in Non-Traditional Vocational Programming (N=1 1 2) VARIABLE B S.E. B Beta Partial R2 R2 Chang_e_ Attitudes About Non- Traditional Work for Female Inmates .116 .053 205 .180 " .148 .148 ** Sexist Attitudes Toward Women -.048 .011 -.381 -.335 ‘* .261 .112 ” Constant 12.312 2.927 * p<.05; *" p<.001 To assess if each of these independent variables significantly contributed to the variance in Scenario One, the partial correlations were examined. The partial correlation for Attitudes About Non- Traditional Work for Female Inmates was significant (Pr2=.180, F=4.80, p=.031). The partial correlation for Sexist Attitudes Toward Women was also significant (Pr’=—.335, F=16.57, p=.000) (see Table 41 ). When examining the variance in Scenario One, both the general attitude about women and a specific attitudinal measure about non-traditional work for female inmates, significantly contributed to whether a respondent would encourage a female inmate to participate in non-traditional vocational 156 programming. Those respondents who were more likely to encourage an inmate to participate in various occupations, regardless of gender, were more likely to encourage a female inmate to participate in non-traditional vocational programming. Furthermore, these individuals with more sexist attitudes were less likely to encourage such participation. To examine if the sex and group membership of the respondent influenced her or his behavioral intentions, an additional regression included these factors. These analyses only examined the correctional officers and program staff Since these two groups included both males and females. In these separate analyses, while the specific attitudinal measure was significant (3:226, F=11.67, p=.002), the general measure, Sexist Attitudes Toward Women did not significantly add to the regression (r2 change=.015., F=.761, p=.388). However, both group membership (r2 change=.078, F=4.31, p=.045) and sex of the respondent (r2 change=.107, F=6.02, p=.012) significantly added to the analyses. Scenario Two This portion of the analysis tested the following research question: Are attitudes about a female inmates' need to work related to behavioral intentions to encourage or discourage her to be financially independent? To further explore this relation, an additional research question was raised: 157 How much of the variation within the behavioral situation of encouraging a female inmate to be financially independent can be explained by attitudes about a female inmate's need to work and how much of the variation can be explained by global attitudes toward women? AS mentioned in Chapter Three, Scenario Two consisted of three scales: a) Financial Independence; b) Work Inside the Home; and 0) Work Outside the Home. The attitudinal scales, Female Inmates' Need to Work and Sexist Attitudes Toward Women, were regressed on each of these three scales. ElnanszlaLIndsnendsnsa The attitudinal measure, Female Inmates' Need to Work did not explain a significant portion of the variance for Scenario Two - Financial Independence (r2=.004, F=0.48, p=.489) (see Table 42). However, Sexist Attitudes Toward Women did explain a significant portion of the variance (r2 change=.128, F change=17.36, p=.0001). Therefore, a respondent's sexist attitudes toward women, rather than the specific attitude about a female inmate's need to work, was a significant predictor as to whether she or he would encourage a female inmate to be financially independent. The more sexist a respondent's attitude, the more likely she or he would discourage a female inmate to be financially independent. The respondent’s sex or group membership did not significantly contribute to the variance. 158 W9 Female Inmates' Need to Work did explain a significant portion of the variance for Scenario Two - Work Inside the Home (F=.086, F=10.98, p=.001 (see Table 42). However, when Sexist Attitudes Toward Women was entered into the analysis, it did not significantly increase the amount of variance (r2 change=.025, F change=3.30, p=.072). In this instance, a respondent's attitude about a female inmate's need to work, rather than her or his sexist attitudes, was a significant predictor of her or his likelihood to encourage a female inmate to work outside the home. Again, the sex and group membership of the respondent did not significantly add to the variance of this behavioral measure. mm For the third scale, Scenario Two - Work Outside the Home, neither Female Inmates' Need to Work or Sexist Attitudes Toward Women significantly explained any of the variance (see Table 42). Sex and group membership also did not significantly explain any of the variance. The best predictor of a respondent's likelihood to encourage a women to be financially independent is his or her sexist attitudes toward women. However, the best predictor of a respondent's likelihood to encourage a women to work outside the home is his or her attitudes about a female inmate's need to work. 159 Scenario Three A multiple regression analysis was conducted to test the following research question: Are attitudes about mothers in prison being "bad" mothers related to behavioral intentions to encourage mothers to retain custody of their children? Table 42 Scenario Two: Financial Situation R2 VARIABLE B S.E. B Beta Partial R2 Change— Financial independence (N=1 21 ) Female Inmates' Need to Work -.085 .058 -.127 -.125 .004 .004 Sexist Attitudes Toward Women .021 .005 .363 .357 ‘ .132 .128 * Constant 2.592 .734 Work Inside the Home (N=1 1 9) Female Inmates' Need to Work .159 .054 .265 .261 '* .086 .086 ** Sexist Attitudes Toward Women .008 .005 .161 .159 .111 .025 Constant 2.317 .672 Work Outside the Home (N=1 20) Female Inmates’ Need .034 .066 .049 .048 .004 .004 to Work Sexist Attitudes Toward .006 .006 .097 .096 .013 .009 Women 4.180 .834 Constant " p<.001; *" p<.01 160 Furthermore, How much of the variation within the behavioral situation of encouraging inmate mothers to retain custody of their children can be explained by attitudes about perceptions of inmate mothers being "bad" mothers and how much of the variation can be explained by global attitudes toward women? ldealizations of Motherhood and Sexist Attitudes Toward Women were entered to assess if these factors could predict a respondent's likelihood to encourage mothers in prison to retain custody of their children (i.e., Scenario Three). ldealizations of Motherhood did not significantly explain any variance in Scenario Three (r2=.032, F=3.82, p=.053) (see Table 43). However, Sexist Attitudes Toward Women Significantly explained a portion of the variance in Scenario Three (r2 change=.041, F change=5.06, p=.026). Although ldealizations of Motherhood did approach significance, when the effects of Sexist Attitudes Toward Women were removed, the partial correlation was -.146 (p=.107). For this Scenario, the more sexist an individual's attitudes toward women, the more likely she or he would discourage a mother to retain custody of her children. When examining the sex and group membership of the participant, neither ldealizations of Motherhood and Sexist Attitudes Toward Women were Significant. The only significant influence in this analysis was group membership (r3change=.137, F change=5.93, p=.020). 1 61 Table 43 Scenario Three: Encourage Mothers in Prison To Retain Custody (1 18) VARIABLE B S.E. B Beta Partial R2 R’ Changei ldeal'zations of Motherhood -.067 .041 -.148 -.146 .032 .032 Sexist Attitudes Toward Women -.013 .006 -.204 -.202 * .073 .041 * Constant -7.784 .819 * p<.05 Scenario Four Again, for this portion of the analysis, two research questions were explored: Are attitudes about inmate mothers maintaining ties with their children related to behavioral intentions to encourage them to maintain ties with their children? Additionally, How much of the variation within the behavioral Situation of encouraging inmate mothers to maintain ties with their children can be explained by attitudes about inmate mothers separation from their children and how much of the variation can be explained by global attitudes toward women? Scenario Four consisted of two scales: a) Encourage Maintaining Ties; and b) Discourage Maintaining Ties. There were two specific attitudinal measures regressed on each of these two scenarios (i.e., Maintaining lies With Children; and Child Visitation Scale) and the global measure, Sexist Attitudes Toward Women. 162 E M'l” II Of the two specific attitudinal measures, only Maintaining Ties WIth Children was a significant predictor of encouraging mothers to maintain ties with their children (p=.021) (see Table 44). Sexist Attitudes Toward Women did not significantly add as a predictor variable (I2 change=.015, F change=2.07, p=.153). Thus, the more positive respondents' attitudes were about mothers maintaining ties with their children, the more likely they would encourage mothers to do so. Sex and group membership did not significantly add to the variance of this behavioral intention. D' llil'i Ii Initially, when Child Visitation and Maintaining Ties Wrth Children were entered, Child Visitation was a significant predictor of discouraging mothers to maintain ties. Sexist Attitudes Toward Women also significantly contributed to the variance (r2 change=.077, F change=10.46, p=.002). However, the results of the partial correlation revealed that when controlling for Sexist Attitudes Toward Women, Child Visitation was not significant. Therefore, the only significant predictor for this scale was Sexist Attitudes Toward Women. In this Situation, a respondent with more sexist attitudes toward women would be more likely to discourage a mother in prison to maintain ties with her children. Again, neither sex nor group membership significantly contributed to the variance. 163 Scenario Five A multiple regression analysis was implemented to address the following two research questions: Are attitudes about female inmates being irrational related to behavioral intentions to interact with inmates in a stressful sfiuafion? Additionally, How much of the variation within the behavioral situation involving interactions with inmates in a stressful situation can be explained by stereotypes about female inmates being irrational and how much of the variation can be explained by global attitudes toward women? Table 44 Scenario Four: Maintaining Ties With Children VARIABLE B S.E. B Beta Partial R’ R2 Chang:- Encourage Maintaining Ties (N=120) Maintaining 'I'Ies with Children .112 .046 .264 209 " .138 .138 ** Child Visitation .103 .097 .119 .091 Sexist Attitudes Toward Women .007 .005 .128 .123 .154 .015 Constant .615 .644 Discourage Maintaining Ties (N=1 20) Female Inmates’ Need .034 .066 .049 .048 .004 .004 to Work Sexist Attitudes Toward .006 .006 .097 .096 .013 .009 Women 4.180 .834 Constant ' p<.05; ** p<.001 164 This analysis was conducted to determine if Female Inmates Are Irrational predict a respondent's likelihood to become involved in an emotional situation between two female inmates (i.e., Scenario Five). The attitudinal measure, Female Inmates Are Irrational, explained a significant portion of Scenario Five variance (12:.047, F=5.67, p=.010) (see Table 45). When Sexist Attitudes Toward Women was entered into the regression, there was a significant increase in the amount of variance explained (r2 change=.063, F=8.04, p=.005). Slightly less than 11% of the variance was explained when regressing Female Inmates Are Irrational and Sexist Attitudes Toward Women on Scenario Five. Table 45 Scenario Five: Involvement in an Emotional Situation (1 1 7) VARIABLE B S.E. B Beta Partial R2 R’ Chang_e_ Female Inmates Are Irrational .201 .082 .215 .215 * .047 .047 * Sexist Attitudes Toward Women .206 .009 .251 251 ” .110 .063 *" Constant .762 1.423 * p<.05; ** p<.01 When examining the variance explained in Scenario Five, both general attitudes about women and the specific attitudinal measure, Female Inmates Are Irrational, significantly contributed to whether a respondent would become 165 involved. A respondent would not get involved in an emotional situation between two female inmates the stronger his or her attitudes that female inmates are irrational. Additionally, respondents were less likely to become involved the more sexist their attitudes toward women. Sex and group membership did not significantly explain any variance in Scenario Five. Scenario Six Finally, the last portion of this analysis examined two research questions: Are attitudes about female inmates being child-like related to behavioral intentions to treat inmates like children? Furthermore, How much of the variation within the behavioral situation involving the treatment of inmates like children can be explained by stereotypes about female inmates being child-like and how much of the variation can be explained by global attitudes toward women? Again, as discussed in Chapter Three, Scenario Six consisted of two scales pertaining to a confrontational situation between two female inmates which could result in some type of paternalistic treatment: a) Personal Involvement; and b) Paternalistic Treatment. The attitudinal scale, Paternalistic Attitudes Toward Female Inmates, and Sexist Attitudes Toward Women were regressed on each of these two scales. 166 W Paternalistic Attitudes Toward Female Inmates did not explain a significant portion of the variance in Scenario Six - Personal Involvement (r2=.009, F=.971, p=.327) (see Table 46). Sexist Attitudes Toward Women also did not significant add any variance to this scenario (1'2 change=.020, F=2.18, p=.143). Thus, respondents' paternalistic attitudes toward female inmates or sexist attitudes toward women were not strong predictors as to whether they would be likely to use some form of paternalistic treatment when there was a confrontational Situation between two inmates. However, when regressing sex and group membership among the correctional officer and program staff, group membership significantly added to the variance in the Personal Involvement Scenario (r2 change=.210, F=10.45, p=.003). William Paternalistic Attitudes Toward Female Inmates was a significant predictor for Scenario Six - Paternalistic Treatment (F=.051, F=6.16, p=.015). However, Sexist Attitudes Toward Women did not significantly explain any additional variance in this Scenario (r2 change=.013, F change=1.53, p=.218). Therefore, the best predictor of respondents' likelihood to treat female inmates in a paternalistic manner are their specific paternalistic attitudes toward female 1 67 inmates and not their sexist attitudes toward women. Sex and group membership did not Significantly add to the variance for this behavioral intention. Table 46 Scenario Six: Paternalistic Situation VARIABLE B S.E. B Beta Partial R2 R2 Change Personal Involvement (N=1 1 0) Paternalistic Attitudes Toward Female Inmates -.115 .088 -.133 -.133 .009 .009 Sexist Attitudes Toward Women .018 .012 .146 .146 .028 .020 Comm 8.849 1 .722 Paternalistic Treatment (N81 1 6) Paternalistic Attitudes Toward Female Inmates .089 .044 .193 .185 * .051 .051 ' Sexist Attitudes Toward Women .008 .006 .1 13 .1 13 .064 .013 Constant 2.441 .868 * p<.05 CHAPTER EIGHT DISCUSSION AND CONCLUSIONS The primary question of this study was "If stereotypes can be found in the criminal justice system and, if the criminal justice system is a mechanism of social control, are stereotypes a form of social control?" From this initial question, two general areas of inquiry were addressed. The first area was to examine if group membership was related to various attitudes concerning female inmates. The second area assessed if these specific attitudinal measures, and one general attitudinal measure, could predict a respondent's behavioral intentions. Specific research questions were drawn from these two areas of inquiry. The conclusions from these inquiries is the main focus of this chapter. The theoretical and policy implications of these findings are also presented. Finally, the limitations of this study as well as recommendations for future research are discussed. Group Membership and Attitudinal Measures To determine if group membership was related to attitudes toward women, analyses of variance were conducted. The results revealed that group membership was significantly related to sexist attitudes toward women. The group with the least sexist attitudes toward women was program staff followed by 168 169 peer counselors and correctional officers. The group with the greatest sexist attitudes toward women was female inmates. To assess if group membership was related to specific attitudes toward female inmates, additional analyses of variance were implemented. Group membership had a significant, independent effect on all but one of these specific attitudinal measures. Two research questions addressed two facets of vocational programming: a) Is group membership related to attitudes about traditional and non-traditional mark for female inmates?; and b) Is group membership related to attitudes about a female inmate's need to work? Two scales were developed to examine these questions: a) Non-Traditional Work for Female Inmates; and b) Female Inmate's Need to Work. Group membership had no significant effect on Attitudes About Non-Traditional Work for Female Inmates. However, female inmates were the least likely to perceive female inmates' need to work compared to the remaining three groups. Program staff had the most positive attitudes concerning female inmates' need to work. Tm research questions examined two aspects of mothers in prison: a) Is group membership related to attitudes about inmate mothers being "bad" mothers?; and b) Is group membership related to attitudes about maintaining ties between inmate mothers and their children? Three scales were administered to address these research questions: a) ldealizations of Motherhood; b) Maintaining Ties with Children; and c) Child Visitation. For all three of these measures, peer counselors had the most positive attitudes concerning mothers 170 in prison followed by female inmates. For two of the three measures, ldealizations of Motherhood and Maintaining Ties with Children, correctional officers had the most negative attitudes followed by program staff. Program staff had the most negative attitudes pertaining to child visitation followed by correctional officers. Thus, prison personnel had the most negative attitudes on these three measures. Two research questions focused on paternalistic attitudes toward female inmates: a) Is group membership related to attitudes about female inmates being irrational?; and b) Is group membership related to attitudes about female inmates being child-like? Two measures were administered to examine these questions: a) Female Inmates are Irrational; and b) Paternalistic Attitudes Toward Female Inmates. For both of these measures, the same pattern emerged among the four groups. Correctional officers had the most negative attitudes followed by program staff. The group with the most positive attitudes was peer counselors followed by female inmates. Again, prison personnel had the most negative attitudes toward female inmates. This portion of the study revealed there were significant differences among the groups concerning various stereotypes of female inmates. The next portion of this study examined if these stereotypes are related to behavioral intentions. 171 Attitudinal Measures and Behavioral Intentions For this section, the results revealed that in some instances the specific attitudinal measure was a better predictor of a respondent's behavioral intentions, and, in other cases, the best predictor was the general measure, Sexist Attitudes Towards Women. Only in two scenario situations was both the specific and general attitude measures significant predictors. I! I' IE . Two research questions were examined to determine if attitudes concerning vocational programming were related to behavioral intentions: Are attitudes about traditional and non-traditional work for female inmates related to behavioral intentions to encourage or discourage an inmate to participate in non-traditional programming? Are attitudes about a female inmate's need to work related to behavioral intentions to encourage or discourage her to be financially independent? Two additional and associated research questions further explored this relation by focusing on how much of the variation is due to specific attitude measures and how much of the variation is due to the general attitude measure (i.e., Sexist Attitudes Toward Women). Both a respondent's specific attitude about non-traditional work for female inmates and sexist attitudes toward women significantly contributed to whether 172 he or she would encourage a female inmate to participate in non-traditional vocational programming. In reference to financial independence, a respondent's sexist attitudes toward women was the best predictor of her or his likelihood to encourage a female inmate to be financially independent. However, a respondent's specific attitude concerning a female inmate's need to work was the best predictor of her or his likelihood to encourage a female inmate to work inside the home. Mil . E' IMII I I Again, two research questions were addressed in order to assess if attitudes about motherhood were related to behavioral intentions: Are attitudes about mothers in prison being "bad" mothers related to behavioral intentions to encourage mothers to retain custody of their children? Are attitudes about inmate mothers maintaining ties with their children related to behavioral intentions to encourage them to maintain ties with their children? As with the previous section, two additional and related questions further examined this association by focusing on the explanatory power of the specific and general attitude measures. For the behavioral situation involving respondents' likelihood to encourage mothers in prison to retain custody of their children, the strongest 173 predictor was their general attitudes about women (i.e., Sexist Attitudes Toward Women) and not their attitudes concerning the idealizations of motherhood. The behavioral situation about maintaining ties with children had two separate, but related, aspects: a) encouraging, or b) discouraging. The best predictor for encouraging female inmates to maintain ties with their children was a respondent's specific attitude about that very issue. However, the best predictor for discouraging a female inmate to maintain ties with her children was her or his general attitudes about women. Eatemaltstislmatment As with the previous sections, two research questions were examined to determine if paternalistic attitudes about female inmates were related to behavioral intentions: Are attitudes about female inmates being irrational related to behavioral intentions to interact with inmates in a stressful sfiuafion? Are attitudes about female inmates being child-like related to behavioral intentions to treat inmates like children? Again, two additional and associated research questions further explored this relation by focusing on how much of the variation was due to the specific, and the general attitudinal measures. Both the specific attitude measure that Female Inmates are Irrational and Sexist Attitudes Toward Women were significant predictors of a respondent's likelihood to become involved in an emotional situation between two female inmates. The best predictor of paternalistic treatment of female inmates is a respondent's paternalistic attitudes toward female inmates rather than his or her general sexist attitudes toward women. Theoretical Implications The theoretical question in this study is whether stereotypes are a form of social control. Specifically, if stereotypes of female inmates exist, is there a potential that such stereotypes are an influential factor in behavior toward female inmates? This behavior is considered a mechanism of social control if it defines as well as limits the behavior of women. Prior to addressing this proposition, possible general factors effecting stereotypes of female inmates were examined. The sex of the respondent had a significant effect on five of the seven attitudinal measures with males having more negative attitudes on four of these scales. Age had a significant correlation on three of these scales. For two of these measures, the older the respondent, the more negative her or his attitudes toward female inmates. Further, when comparing respondents' attitudes toward male and female inmates, there were significant differences on four of the six measures. Three of these four scales pertained to parents in prison. Interestingly, mothers in prison were perceived more positively compared to fathers in prison. However, 174 175 respondents had more paternalistic attitudes toward female inmates when compared to attitudes toward male inmates. In order to examine the above proposition, it was essential to initially determine if such stereotypes of female inmates do exist among various groups within the correctional facility. It was also important to assess that if such group differences exist, was this due to group membership alone, or could these differences be explained by social distance? Social distance significantly differed among the various groups. However, social distance and group membership never Significantly effected any of the attitudinal measures simultaneously. The results of this study revealed that stereotypes of female inmates significantly differed among the groups on six of the seven attitudinal measures. Social distance significantly effected two of the seven scales. The final portion of this study then focused specifically on whether these stereotypes were related to behavioral intentions. As discussed in a previous chapter, behavioral intentions are the immediate determinant of an individual's overt behavior. The results of this research revealed that stereotypes did have a significant impact on behavioral intentions. In some instances, general stereotypes of women had more of a significant influence on behavioral intentions while in other instances stereotypes pertaining specifically to female inmates had more of an impact. As stated previously, the initial question of this research was "If stereotypes can be found in the criminal justice system, and if the criminal justice 176 system is a mechanism of social control, are stereotypes a form of social control over female offenders?" First, this study empirically demonstrated that stereotypes of female inmates exist, and differ, among the various groups within the prison. Second, these stereotypes, either specific attitudes about female inmates or general attitudes about women, were significantly related to behavioral intentions. As mentioned previously, such behavioral intentions are mechanisms of social control if they define as well as limit the behavior of female inmates. There were cases in which this association was illustrated, specifically in the behavioral intentions concerning mothers in prison and paternalistic treatment. These negative stereotypes of female inmates among correctional personnel supports previous feminist criticisms concerning the experiences of women in prison (e.g., Pollock-Byme, 1990; Rafter, 1990). Furthermore, such negative stereotypes were significantly related to relevant behavioral intentions. Therefore, the theoretical perspective of stereotypes as a form of social control has clearly been supported in cases involving mothers in prison and paternalistic treatment. The results of this study also revealed, however, that there was not a clear, decisive association between those in a powerful position (i.e., correctional personnel) exerting more social control over female inmates in the area of vocational programming. The group with the strongest negative stereotypes was among female inmates and not correctional personnel. 177 Furthermore, female inmates had the most sexist attitudes toward women, followed by correctional officers. As mentioned previously, these stereotypes did have a significant impact on some of the behavioral intentions. It is essential to emphasize, however, that feminist criticisms of vocational programming for women in prison has primarily focused on more organizational or institutional levels (e.g., Carlen, 1982; Carp & Schade, 1993; Chapman, 1980; Morash, Haarr, & Rucker, 1994; Meyer, 1984; Simon & Landis, 1991; Weisheit, 1984). This study emphasized issues pertaining to vocational programming on an individual level. Thus, this study can provide a "qualified" answer to the above mentioned question as to whether stereotypes are a form of social control over female inmates. However, this answer itself raises more questions which are beyond the scope of this study. For instance, are these negative stereotypes among female inmates exacerbated by their prison experience or do female inmates, prior to prison, have such stereotypes regardless of their length of incarceration? Policy and Program Implications AS discussed in Chapter Two, Heidensohn (1985) argued, "[o]f all the subtler constraints on the way women act and are supposed to act, few are more complex than the workings of social policies (p. 191). Policy can be influenced, in a subtle as well as blatant manner, by stereotypes. If various groups within a correctional facility have negative stereotypes concerning women in general, 178 and specifically female inmates, this could be related to subsequent interactions with female inmates. This study has illustrated that stereotypes of female inmates do differ by groups. Also, these stereotypes were associated with negative behavioral intentions. Again, female inmates had the most sexist attitudes toward women and they were the least likely to perceive female inmates' need to work. Thus, it is essential to recognize that while vocational programs may be implemented, female inmates' participation and success in such programs may be contingent on their overall attitudes toward such programming. Education, such as women's studies courses, may be a useful approach for female inmates to participate and succeed in such programming opportunities. Essentially, just implementing programming may not be sufficient to ensure female inmates have ample opportunities to succeed. Rather, additional issues need to be addressed along with program opportunities. On numerous attitudinal measures, peer counselors had positive attitudes toward female inmates. This group of individuals currently provide services to the other female inmates. However, these individuals could also enhance possible feminist programming or education within the facility. Furthermore, peer counselors had the highest score on the Social Distance Scale - Female Inmates (i.e., they would be more likely to be involved in personal interactions with female inmates). Thus, they have already demonstrated a relationship, or willingness, to interact with other female inmates. 179 There were also group differences among correctional officers and program staff concerning inmates. One explanation is that each group interacts differently with female inmates due to their respective duties and responsibilities. This was demonstrated in the Social Distance Scale - Female Inmates. After peer counselors, program staff had the next highest score. However, correctional officers had the lowest score among the other three groups. This raises the issue concerning a lack of consistency as to how female inmates are treated - not due to their offender status but rather their sex. Furthermore, Since Specific as well as general sexist attitudes were related to various behavioral intentions, both of these attitudinal viewpoints could be addressed during recruitment, training, and supervision. The literature has suggested that female inmates have different needs. Therefore, it is essential to assess if training, supervision, and overall departmental policies have incorporated these special needs of female inmates, or are such policies primarily based on male-inmate correctional facilities. For instance, if correctional personnel have been trained in handling violent, confrontational situations, when those situations occur, they will be able to draw on that training to manage such a situation. However, if correctional personnel have not been trained on how to handle situations which are emotional in nature, than they may be frustrated due to a lack of training as to how to manage such a situation. 180 Limitations and Recommendations The primary limitation of this study is the generalizability of these findings to other correctional facilities for women. There are factors unique to Bedford Hills that are not characteristic of other facilities. One example is the group of female inmates who also provide some time of services (i.e., peer counselors). Another unique factor of Bedford Hills is the number of program opportunities for female inmates (e.g., AIDS Counseling and Education; Mothers and Children Program). Also, Bedford Hills is located approximately one hour away from New York City. Therefore, there is limited generalizability to facilities located in a characteristically more rural area. However, this limited generalizability essentially applies to correctional officers and program staff. The sample of female inmates in this study was similar to the national profile of female inmates (BJS, 1994). One recommendation for future research would be to replicate this study in various correctional facilities differing by such factors as regional location and available programming opportunities. Thereafter, these additional factors could be included to assess if organizational factors could possibly be influential variables. Another limitation of this study was the low number of correctional officers in the sample. As stated previously, this group was initially to be designated as "newl' and "senior" correctional officers. Due to the low response rate, these two groups, however, had to be collapsed into one. One 181 recommendation would be to include a correctional officer, working within the facility, as a contact person. This individual could provide the assurance to correctional officers that is otherwise difficult for an "outsider" to provide them. Finally, one additional recommendation would be to replicate this study in an all-male facility. Comparisons could be made between the responses of female inmates and male inmates and their experiences. This study could also include more in-depth interviews to obtain more qualitative types of data. APPENDICES APPENDIX A 182 Appendix A CONSENT FORM Case Name: Number (PLEASE PRINT) The purpose of this research is to obtain information about various viewpoints concerning female inmates. This is a ”paper and pencil” survey which will be conducted in one session. I understand met my participation in this research is completely voluntary and of my free-will. I can choose to refuse participation at any time while taking this survey. There are no consequences for not participating in this research. I understand that my responses are confidential. In other words, I understand that the researcher will be the 9.0]! person who knows my real name. No person associated with the Department of Corrections is involved in this process. I understand that to ensure this confidentiality, after completing this test, this page will be removed from the test. If I have any quedons about my participation in this project, I may contact Pamela Schram Graduate Student School of Criminal Justice Mchigan State University 560 Baker Hall East Lansing, Michigan 48824 Date: Signature: APPENDIX B 183 Appendix B Case Number PART I Instructions to Respondents In the first part of this survey, I would like to ask you some questions about your background. Remember that your participation in this survey is completely voluntary and confidential. No one will know your name except for myself. If at any time you would like me to read the questions or you are not sure what I am asking, please let me know. 1. How old are you? years old 2 Are you A) African American 8) Hispanic C) White D) Asian American E) Native American F) Other (please explain) 3. Below, list how many children you have, if any, starting with the oldest child. Please indicate step-child with *. Birth Living at Sex Date Home? (Circle) Month Year Yes or No Child 1 M F Y N Child 2 M F Y N Child 3 M F Y N Child 4 M F Y N Child 5 M F Y N Child 6 M F Y N Child 7 M F Y N Child 8 M F Y N Child 9 M F Y N Child 10 M F Y N If more than 10 children/step-children, please check here 1 84 How many years of school have you completed? Up to 8th grade 9th grade 10th grade 11th grade High school graduate GED One year of college Two years of college Three years of college Bachelors degree Frve years or more of college Other (please explain below) FFFFIQWWPPWP Have you had any vocational training? A. Yes 8. No If yes, what kind of vocational training? When did you receive this vocational training? A. while incarcerated 8. while on parole or probation C. while on the "outside" D. other (please explain below) Immediately before you were locked up, whether in prison or jail, what was your primary source of income? (Check one only) Job State Assistance Ge, AFDC, food stamps) Family Friends Combination of the above (please explain below) Other (please explain below) mmpow> 10. 11. 12. 13. 185 Were you employed before you were incarcerated? A. Yes B. No C. OfiIer (please explain) If you were employed immediately before you were locked up. were you A. Full time (40 hours) B. Part time (at least 20 hours) C. Partially (from 1 to 19 hours) D. Not employed E. Other (please explain below) Please fill out the following table. What is the current offense(s) for which Length of you are in prison? sentence When were you admitted into prison? How many times have you been previously incarcerated? APPENDIX C 186 Appendix C Case Number PART I Instructions to Respondents In the first part of this survey, I would like to ask you some questions about your background. Remember that your participation in this survey is completely voluntary and confidential. No one will know your name except for myself. If at any time you would like me to read the questions or you are not sure what I am asking, please let me know. 1. How old are you? years old 2 Are you A) African American 8) Hispanic C) White D) Asian American E) Native American F) Other (please explain) 3. Below, list how many children you have, if any, starting with the oldest child. Please indicate step-child with '. Birth Living at Sex Date Home? (Circle) Month Year Yes or No Child 1 M F Y N Child 2 M F Y N Child 3 M F Y N Child 4 M F Y N Child 5 M F Y N Child 6 M F Y N Child 7 M F Y N Child 8 M F Y N Child 9 M F Y N Child 10 M F Y N If more than 10 children/stepchildren, please check here $7 How many years of school have you completed? Up to 8th grade 9th grade 10th grade 11th grade High school graduate GED One year of college Two years of college Three years of college Bachelors degree Frve years or more of college Other (please explain below) FZFFIQWWPOP? Have you had any vocational training? A. Yes 8. No If yes, what kind of vocational training? When did you receive this vocational training? while incarcerated while on parole or probation while on the "outside” other (please explain below) new? Immediately before you were locked up, whether in prison or jail, what was your primary source of income? (Check one only) Job State Assidance (i.e., AFDC, food stamps) Family Friends Combination of the above (please explain below) Other (please explain below) TIFF-3.0!”? 10. 11. 12. 13. 1 88 Were you employed before you were incarcerated? A. Yes B. No C. Other (please explain) If you were employed immediately before you were locked up, were you A. Full time (40 hours) B. Part time (at least 20 hours) C. Partially (from 1 to 19 hours) D. Not employed E. Other (please explain below) Please fill out the following table. What is the current offense(s) for which Length of you are in prison? sentence When were you admitted into prison? How many times have you been previously incarcerated? 189 14. What are your responsibilities as a peer counselor? 15. How long have you been a peer counselor? APPENDIX E Instructions to Respondents 190 Appendix D Case Number PART I In the first part of this survey, I would like to ask you some questions about your background. Remember that your participation in this survey is completely voluntary and confidential. No one will knowyour name earceptfor myself. Be sure to stop and ask me any queflonsthatyou might have as you fill them out 1. How old are you? years old 2 Are you A) African American 6) Hispanic C) White D) Asian American E) Native American F) Other (please emlain) 3. Are you female or male? A. Female 8. Male 4. How many years of scth have you completed? Up to 8th grade 9th grade 10th grade 11th grade High school graduate GED One year of college Two years of college Three years of college Bachelors degree Fee'sewmppwr Five years or more of college Other (please explain below) 193 How long have you been employed as a program staff member? Years Months How long have you been employed as a program staff member at this facility? (include prior assignment to work at Bedford Hills) Years _ Months Have you ever been employed as a program staff member in a male correctional facility? A. Yes 8. No C. Other (please explain below) If yes, how long were you employed? (Include all prior assignment) Years Months What types of services do you currently provide to the female inmates? APPENDIX F 194 Appendix F Below are seven types of interactions that could occur between people on a day-to-day basis. For each interaction, give your first reaction for each of the four groups of people. For example, if you ”Would say 'hi' to a new correctional officer,” then place a check mark in the box. EXAMPLE: New SenIor Interactions Inmates Counselors Staff when ou saw lfyouwould notsay'li,’then|agygithhnk. Do notg'veyourreactionstothe bedortheworst people in that group. Inwad, your responses should be based on your general, overall reaction to individuals within that group. New Senior Female Peer I Correctional Correctional Program Interactions Inmates Counselors I (Officers Staff Would you avoid Would you nod in acknowledgement when you saw Would you smile when you saw Would you say 'hi' to Would you briefly talk to Would you sincerely ask, “How are you doing?“ to Would you talk about your personal ,, oblems to APPENDIX G 1 95 Appendix G PART II Instructions For the statement listed below, please circle to response that best reflect your opinion. There are no right or wrong answers, only opinions. You are asked to express your feelings about each statement by indicating whether you strongly agree, moderately agree, slighfly agree, neither agree or disagree, slightly disagree, moderately disagree, and strongly disagree. Response Choices 1 = Strongly agree 4. Neither agree or disagree 2 = Moderately agree 5. Slightly disagree 3 = Slightly agree 6. Moderately disagree 7. Strongly disagree 1. If I had a daughter, I would discourage her from working on cars. 1 2 3 4 5 6 7 2. I get angry at women who complain that American society is 1 2 3 4 5 6 7 unfair to them. 3. Our society put too much emphasis on beauty, especially for 1 2 3 4 5 6 7 women. 4. Women shop more than men because they can‘t decide what 1 2 3 4 5 6 7 to buy. 5. Most feminist are hopping on the bandwagon of protest just 1 2 3 4 5 6 7 for the fun of it 6. ltbothersmewhenamanisinterestedinawomanonlyifshe 1234567 is pretty. 7. Itbothersmetoseeamanbeingtoldwhattodobyawoman. 1234567 8. I think that having children is a woman's greatest fulfillment. 1 2 3 4 5 6 7 9. Men are instinctually more courageous than women in die face of 1 2 3 4 5 6 7 danger. 10. lthinkthatwomenshouldspendalotoftimetryingtobepretty. 1234567 11. I can really understand why there needs to be a women's 1 2 3 4 5 6 7 right movement. 12. Women rely more on intuition and less on reason than men do. 1 2 3 4 5 6 7 196 Response Choices 1 = Strongly agree 4. Neither agree or disagree 2 = Moderately agree 5. Slightly disagree 3 = Slightly agree 6. Moderately disagree 7. Strongly disagree 13. Women should not be as sexually active before marriage as men. 1 2 3 4 5 6 7 14. Men are justaseadly influenced byothersaswomen are. 1 23456 7 15. I think women should be more concerned about their appearance 1 2 3 4 5 6 7 than men. 16. Men will always be the dominant sex. 1 2 3 4 5 6 7 17. I dislike it when men treat women as sexual object. 1 2 3 4 5 6 7 18. I think that the husband should have the final say when a couple 1 2 3 4 5 6 7 makes a decision. 19. Women should have all the same right as men. 1 2 3 4 5 6 7 20. I see nothing wrong with a woman who doesn‘t like to wear 1 2 3 4 5 6 7 skirt or dresses. 21. Women should be handled gently by men because they are so 1 2 3 4 5 6 7 delicate. 22. Women should be prepared to oppose men In order to obtain 1 2 3 4 5 6 7 equal status. 23. I am suspicious of a woman who would rather work than have 1 2 3 4 5 6 7 children. 24. I think that women are naturally emotionally weaker than men. 1 2 3 4 5 6 7 25. On the average, women are as intelligent as men. 1 2 3 4 5 6 7 26. If a husband and wife both work full time, the husband should 1 2 3 4 5 6 7 do half of the housework. 27. I like women who are outpoken. 1 2 3 4 5 6 7 28. I see nothing wrong with men whistling at shapely women. 1 2 3 4 5 6 7 29. ltbothersmemoretoseeawomanwhoispushythanaman 1234567 who is pushy. 197 Response Choices 1 = Strongly agree 4. Neither agree or disagree 2 = Moderately agree 5. Slightly disagree 3 = Slightly agree 6. Moderately disagree 7. Strongly disagree 30. A working wife should not be hired for a job ifthere is a 2 3 4 5 6 7 family man who needs it. 31. Women can handle pressure just as well as men can when 1 2 3 4 5 6 7 making a decision. 32. Men are naturally better than women at mechanical things. 1 2 3 4 5 6 7 33. A woman's place is in the home. 1 2 3 4 5 6 7 34. I think that many TV commercials present a degrading picture 1 2 3 4 5 6 7 of women. 35. I think a woman could do most things as well as a man. 1 2 3 4 5 6 7 36. I think that men are instinctually more competitive than women. 1 2 3 4 5 6 7 37. I think women have a right to be angry when they are referred 1 2 3 4 5 6 7 to as a "broad.” 38. It would make me feel awkward to address a woman as “Ms." 1 2 3 4 5 6 7 39. I see nothing wrong with men who are primarily interested in 1 2 3 4 5 6 7 a woman's body. 40 1 2 3 4 5 6 7 lfl had a choice, lwouldjustas soon work for a woman asforaman. N APPENDIX I-l 1 98 Appendix H Below, I am going to list some different jobs. In each case, would you tell me if you would a) encourage a male inmate for thatjob, b) encourage a female inmate for that job, or c) encourage both a male and female inmate for that job. Please check the box that best reflect your opinion. Remember, there are no right or wrong answers. — M _ "A“: °'* FEMALE INMATE FEMALE INMATE Autobody repair Brick masonry Buildino maintenance Carpentry Clerical CosmetoEgy Data processing Dental technician Draflin 9 Electronics Engine and appliance repair Food service . Home economics Housekeeping Laundry Machine shop Metal work Nurse's aide Plumbin Receptionist aide T rile WOIdI 9 APPENDIX I 199 Appendix I Below are some questions about tmale inmates. The response choices are as follows: SA - Strongly Agree; A -— Agree; UD - Undecided; D — Disagree; and SD - Strongly Disagree. Remember - there are no right or wrong answers. Circle the response which best reflect your opinions. Be sure to stop and ask me any questions that you might have as you fill them out. Response Choices SA Strongly Agree D Disagree A Agree SD Strongly Disagree UD Undecided ITEM FEMALE INMATE If a female inmate has children, upon release, she will most likely be getting money to support them. Many mothers in prison have a hard time bei 9 oood parent to their children. It is always important for children to visit their mothers in prison. Women in prison are usually too emotional. Women in prison are as responsible as other adult. Upon release, mod female inmates will need to be financially independent on their own income. Even though a mother is in prison, it does not mean she is a bad parent Children should not be exposed to seeing their mothers in prison. Women in prison are usually irrational. Women in prison are a lot like children. Many women in prison can find someone, such as friends or family, to help them financiall when released. 200 Response Choices SA Strongly Agree 0 Disagree A Agree SD Strongly Disagree UD Undecided ITEM FEMALE INMATE Upon release, most female inmates will be oood mothers to their children. UD D It is important for a mother to be reunited with her children upon release. UD D Often women in prison need to be treated like children rather than adult. UD Many female inmates who get out of prison really do not want to work on a regular hm. People who are incarcerated tend to have some problems being_good parent. Children should have just limited contact with their mothers while they are incarceratd. Some women in prison throw ”temper tantrums.” Upon release, a female inmate will need to combine parentingand a job. Most female inmates were not very good mothers prior to beinglocked up. Women in prison m their children while incarcerated. Women in prison are usually rational. Women in prison should be treated like adult. Upon release, female inmates usually have the option of working outide the home or getting financial support so they can care for fami members. 201 Response Choices SA Strongly Agree D Disagree A Agree SD Strongly Disagree UD Undecided ITEM FEMALE INMATE lfwomen in prison had problems being good mothers to their children prior to incarceration, most of them have addressed these problems while in prison. Women in prison regret being separated from their children due to their incarceration. Women in prison can rationally express their a er. Women in prison will have a significant other in their life that will help them financially when released from prison. Children of most female inmates should be given up for mat custody to others such as relatives so the children can have some stabilityin their lives. Maintaining contact between mothers in prison and their children helps them reunite u . . n release. Women in prison have a tendency to overreact. Many women in prison complain too much. The vast majority of mothers in prison need to learn the basics of howto be an adequate parent Children need to see their mothers when their mothers are in prison. In most instances, when a conflict occurs between two female inmates, it is easy to calml resolve the Situation. 202 Response Choices SA Strongly Agree D Disagree A Agree SD Strongly Disagree UD Undecided Many women in prison are weak. Many women in prison can reasonably deal with disa. . . 'ntrnent FEMALE INMATE D SD D SD APPENDIX J 203 Appendix J Below are some questons about male inmates. The response choices are as follows: SA - Strongly Agree; A — Agree; UD — Undecided; D - Disagree; and SD - Strongly Disagree. Remember - there are no right or wrong answers. Circle the response which best reflect your opinions. Be sure to stop and ask me any queflons that you might have as you fill them out. Response Choices SA Strongly Agree D Disagree A Agree SD Strongly Disagree UD Undecided ITEM FEMALE INMATE If a male inmate has children, upon release, he will most likely be getting money to support them. Many fathers in prison have a hard time bei 9 . ood parent to their children. It is always important for children to visit their , fathers in glam. Men in prison are usually too emotional. Men in prison are as responsible as other adult. Upon release, most male inmates will need to be financially independent on their own income. Even though a father is in prison, it does not mean he is a bad parent Children should not be exposed to seeing their fathers in Eison. Men in prison are usually irrational. Men in prison are a lot like children. Many men in prison can find someone, such as friends or family, to help them financially when released. 204 Response Choices SA Strongly Agree D Disagree A Agree SD Strongly Disagree UD Undecided Upon release, most male inmates will be 9 . - . fathers to their children. It is important for a father to be reunited with his children upon release. Often men in prison need to be treated like children rather than adult. Many male inmates who get out of prison really do not want to work on a regular basis. People who are incarcerated tend to have some problems beinggood parent. Children should have just limited contact with their fathers while they are incarceratd. . Some men in prison throw “temper tantrums.” Upon release, a male inmate will need to combine parentimand ajob. Most male inmates were not very good fathers prior to being locked up. Men in prison mts their children while incarcerated. Men in prison are usually rational. Men in prison should be treated like adult. SA SA SA SA SA SA SA Upon release, male inmates usually have the option of working outide the home or golfing financial support so they can care for family members. Response Choices SA Strongly Agree D Disagree A Agree SD Strongly Disagree UD Undecided If men in prison had problems being good fathers to their children prior to incarceration, most of them have addressed these . oblems while in prison. Men in prison regret being separated from their children due to their incarceration. Men in prison can rationally express hair a er. Men in prison will have a significant other in their life that will help them financially when released from prison. Children of most male inmates should be given up for permanent custody to others such as relatives so the children can have some stability in their lives. Maintaining contact between fathers in prison and their children helps them reunite upon release. Men in prison have a tendency to overreact. Many men in prison complain too much. The vast majority of fathers in prison need to learn the basics of how to be an adequate parent Children need to see their fathers when their fathers are in prison. In most instances, when a conflict occurs between tvro male inmates, it is easy to calml resolve the Situation. 206 Response Choices SA Strongly Agree D Disagree A Agree SD Strongly Disagree UD Undecided Many men in prison are weak. FEMALE INMATE D SD Many men in prison can reasonably deal with APPENDIX K 207 Appendix K Below are six scenarios or stories. After each scenario are a few questions asking how you might respond in that situation. The response choices are as follows: VL — Very likely; SL — Somewhat likely; NL - Not very likely; and NA - Not at all. Remember — there are no right or wrong answers. Circle the response which best reflect your opinions. Be sure to stop and ask me any questions that you might have as you fill them out. 551nm Susan was a twenty-four year old single mother of two children. She was convicted of larceny and serving a two year sentence. She wanted to pursue a job as a plumber. She was going to enter a vocational program for this type of occupation. She came to you to ask for your advice on this job choice. Response Choices VL Very Likely NL Not Very Likely SL Somewhat Likely NA Not At All “ nespousecuouces I . How likely would you tell Susan that it will be difficult for her to obtain this type of job? How likely would you tell Susan that it is important for her to select a job that is suited specifically for a woman's unique abilities? How likely would you encourage Susan to rticipate in the progfln? How likely would you tell Susan that she should . obably reconsider her decision? 1 How likely would you tell Susan that she should not - inthe . . oram? APPENDIX L 208 Appendix L W Andrea was a twenty-six year old mother ofthree. She was serving a three year sentence for breaking and entering. Upon release, Andrea wanted to stay home to raise her children. She was hoping that when she was released she would marry her boyfriend of four years. He would financially support her aswell as her children. She cametoyoutoaskforyouradvice onthisidea. Response Choices VL Very Likely NL Not Very Likely SL Somewhat Likely NA Not At All “ Responsecuonces How likely would you tell Andrea that it is important for her to be financially independent on her own income? How likely would you tell Andrea that she should pursue some type of occupation outide the home? How likely would you suggest that Andrea participate in a vocational program in case she needs to obtain employment in the future? How likely would you encourage Andrea to stay at home and raise her children? How likely would you encourage Andrea to work in the home while her boyfriend provides financial support for the family? How likely would you tell Andrea not to stay at home and raise her children? How likely would you tell Andrea not to work in the home full time while her boyfriend provides financial su y . rt for the fam' APPENDIX M 209 Appendix M W Cassandra has two children. Prior to being locked up, she had difficulty taking care of her children mcause ofheraddiclion to cocaine. On many occasions, her mother would take care ofhe kids while Cassandra was "on the street.” While incarcerated, Cassandra's moher had temporary custody of he children. Cassandra has taken substnce abuse counseling and participates weekly in NA and AA meetings. Cassandra has just been informed that her moher want permanent custody of the chidren. Cassandra is not sure if she should let her mother keep the children or not. She came to you to ask for your advice on his situation. Response Choices VL Very Likely NL Not Very Likely SL Somewhat Likely NA Not At All “ nesvousecuovcss How likely would you tell Cassandra hat she had difficulty taking care of her children when she was on he "outide?” How likely would you tell Cassandra hat she has made positive steps (e.g., substance abuse counseling) in order to take care of her children? How likely would you tell Cassandra hat she would, without a doubt, be a good mother to her children upon release? How likely would you encourage Cassandra to let her moher have permanent custody? How likely would you tell Cassandra not to let her moher have permanent custody? How likely would you tell Cassandra hat he children would "be better off” wih her moher? APPENDIX N ‘J x 210 Appendix N W Rachel is a mother of hree children. She is currently serving a hree year sentence for larceny. Rachel‘s sister, Karen, is taking care of her children. Rachel received a letter from Karen explaining that Rachel should not have any contact with her children while incarcerated. Karen argued hat seeing heir moher in prison could negatively effect he children. Rachel came to you to ask for your advice about his situation. Response Choices VL Very Likely NL Not Very Likely SL Somewhat Likely NA Not At All “ RESPONSEcuorcEs How likely would you tell Rachel hat it is important to maintain contact wih her children? How likely would you tell Rachel hat it is difficult for he children to see her in prison? How likely would you tell Rachel hat it would be better for her children ifhey did not see her in prison? How likely would you encourage Rachel not to have any contact with her children while in o . o I]? How likely would you encourage Rachel to continue having contact with her children while in prison? How likely would you avoid talking to Rachel his to -‘ ? . APPENDIX 0 21 1 Appendix 0 5 'E' Shannon was an inmate at he Corbin Women's Correctional Facility. She had just received some news from her famiyhatwas qtie upsetting. She was very emotional and was not rationally thinking hrough he siuation. She was crying quite loudly and physically shaking. You walked in the unit and saw Shannon. Response Choices VL Very Likely NL Not Very Likely SL Somewhat Likely NA Not At All -'I'I!_ Remus-scams How likely would you avoid talking to Shannon? How likely would you ask Shannon what was How likely would you attempt to comfort Shannon? How likely would you suggest hat Shannon talk to someone? How likely would you sit down and talk to Shannon? How likely would you tell Shannon that she should not react to situations in his manner? APPENDIX P 212 Appendix P W Sandy and Karla lived in he same unit. Sandy kept her area neat and organized while Karla was very unorganized and somewhat ”sloppy." This always caused some mild tension between he two. One day after her parole hearing, Sandy came back to her unit. She was quite upset because he parole board just gave her a 12 month “hit at he board.” Sandy noticed hat Karla left some of her dirty clothes on Sandy's bed. Sandy got very upset and soon Sandy and Karla were arguing. Instead of resolving his situation in an adult fashion, boh inmates were acting in a childish manner. You were in he unit when hey were arguing. Response Choices VL Very Likely NL Not Very Likely SL Somewhat Likely NA Not At All “ Response cuouces How likely would you avoid he situation? VL SL NL NA How likely would you ask Sandy and Karla what he were an ui ., about? VL SL NL NA How likely would you try to calmly talk to Sandy and Karla? VL SL NL NA How likely would you take any formal action? VL SL NL NA How likely would you attempt to settle he a eument? VL SL NL NA How likely would you tell hem hat hey were self 9 like children? VL SL NL NA How likely would you feel as hough you would have to handle the sihialion as hough you were scoldi . children? I WOULD LIKE TO THANK YOU FOR TAKING THE TIME TO COMPLETE THIS SURVEY. YOUR PARTICIPATION IS GREATLY APPRECIATEDII LIST OF REFERENCES LIST OF REFERENCES Abelson, R. P. (1982). Three modes of attitude-behavior consistency. In M. P. Zanna, E..T Higgins, 80. P. Herman, (Eds.,) W Behaxmlhaflntaflsfixmmsm pp 131-146 Hillsdale NJ Lawrence Erlbaum Associates. Ajzen, l. (1982). On behaving In accordance with one' s attitudes. In M. P. Zanna, E..T Higgins, &C. P. Herman (Eds) W W. Vol 2 PP 3-15 Hillsdale NJ Lawrence Erlbaum Associates. Ajzen, l. (1989). Attitude structure and behavior. In AR. Pratkanis, S.J. Breckler. & AG. Greenwald (Eds). Wigs pp. 241 -269. Hillsdale, NJ: Lawrence Erlbaum Associates. Allport, G.W. (1935). Attitudes. In C. Murchin, (Ed), MW Esychglggy. Worchester, MA: Clark University Press. Amir, D. & Biniamin, O. (1991). Abortion approval as a ritual of symbolic control Mmenanslfiriminauustice. 3 5- 25 Ashmore, R. D., & Del Boca, F. K (1981). Conceptual approaches to stereotypes and stereotyping. In D. L. Hamilton (Ed), Cognitive 2 . ,pp.1-35. Hillsdale, NJ. Lawrence Erlbaum AsSociates. Babbie, E. (1990). W. (2nd Ed.). Belmont, CA: Wadsworth Publishing Company. Bagozzi, R. P. & Bumkrant, R. E. (1979). Attitude organization and the attitude- behavior relationship WM .43. 47-57. Baunauch, P.J. (1985). Motheminprisgn. New Brunswick, NJ: Transaction, Inc. Beckerman, A. (1991). Women in prison: The conflict between confinement and parental rights. W. 18, 171-183. Bedford Hills Correctional Facility. (1994). WIS QanstianaLEactlI’tv Bedford Hills NY 213 214 Belknap. J. (1996). W Belmont, CA: Wadsworth Publishing Company. Benson, P.L., & Vincent, 8. (1980). Development and validation of the Sexist Attitudes Toward Women Scale (SATWS). W W. 5. 276-291. Biemat, M. & Crandall, C. S. (1994). Stereotyping and contact with social groups: Measurement and conceptual issues. W W. 2.4. 659-677 Bogardus, ES. (1925). Social distance and its origins. Smiolggyandfiogial 33591911. 3. 216-225. Bogardus, E. S. (1926). Social distances between groups. W Sogiologx.1_0.473-479 Bogardus, ES. (1933). Asocial distance scale. Sgciglggyandfiggial Research. 22. 265-271. Bohmstedt. G.W.. &Knoke. D. (1988). W: (2d Ed.). Itasca,le F.E. Peacock Publishers, Inc. Bowker. LH. (1978). W. Lexington, MA: 0.0. Heath and Company. Branscombe, NR. 8. Smith, ER. (1990). Gender and racial stereotypes in impression formation and social decision-making processes. W. 22, 627-647. Brewer, M. B, & Kramer, R. M. ( 1985) The Psychology of lntergroup Attitudes and Behavior AnnuaLBexiemEmholoox. E. 219-243 Brinkerhoff, M. B. & Jacob, J. C. (1994). Racial, ethnic and religious social distance In Surinam: An exploration of the 'strategic alliance hypothesis' in a Caribbean community. WM, 11, 636-661. Brownmiller, S. (1984). Eemininitx. New York: Fawcett Columbine. Bureau of Justice Statistics. (March, 1991). Women in prison. US Department of Justice. 215 Bureau of Justice Statistics. (March, 1994). Women in prison. US Department of Justice. Burkhardt, K (1973). flomonjnpnaon. Garden City, NJ: Doubleday. Carlen, P. (1982). Papa' s discipline: An analysis of disciplinary modes In the Scottish women' 3 prison. W m, 97-124. Carp, S. V. & Schade, L. S. (1993). Tailoring facility programming to suit female offenders needs. In : - - - Population, American Correctional Association. Chapman. J. (1980). WWW Lexington. MA: Lexington Books. Chesney-Lind, M. (1991). Patriarchy, prisons, and jails. A critical look at trends in women's incarceration. W, 11, 51 -67. Chesney-Lind, M. (1986). Women and crime: The female offender. Signo,_ JoumaloflflomeanfiultumndSooiem. 12. 78-96 Chesney-Lind, M. (1978). "" In L. H. Bowker, Wm iustmaxstom, pp. Lexington, MA: D. C Heath and Company. Chunn, D.E. & Gavigan, SA (1988). Social control: Analytic tool oranalytic quagmire. W 12.107-124. Cohen, S. (1983). Social control talk: Telling stories about correctional change. In D. Garland 8. P. Young, (Eds.),]1]oPMLto_punian, pp. 101- 129. London: Heinemann. Cohen, s. (1985). MW QIasoifloation. Cambridge. Polity Press. Datesman, S.K & Cales, G.L. (1983). "I'm still the same mommY": Maintaining the mother/child relationship in prison. Inopnsonjoumal, Q3, 142-154. Davin, A. (1978). Imperialism and motherhood. HistonLlnLongoIJop, 5, 9-65. DeVelliS. RF. (1991). WWW Newbury Park, CA: Sage Publications. 216 Dobash, R.P., Dobash, R.E., & Gutteridge, S. (1986). Inoimpnsonmontot man. Oxford: Basil Blackwell, Ltd. Dugger, K (1991). Social location and gender-role attitudes. A comparison of black and white women. In J. Lorber and S A Farrell, (Eds. ), Ina__sooial constmctionotosnder. pp 38-59 Newbury Park CA: Sage Publications. Dunn-Rankin, P. (1983). W. Hillsdale, NJ: Lawrence Erlbaum Associates. Edwards, AR. (1989). Sex/gender, sexism and criminal justice: Some theoretical considerations WWW. 11,165-184. Etaugh, C., & Spiller, B. (1989). Attitudes toward women: Comparison of traditional-aged and older college students. Joumalotgollogofituoont anelooment. 33. 41-46. Etzioni, A. (1961). Fairchild, H.H. (1985). Black, Negro, or Afro-American? The differences are crucial! W. 13. 47-55. Feinman, C. (1983). An historical overview of the treatment of incarcerated women: Myths and realities of rehabilitation. IboPflsonJoumal, 53, 12- 26. Feinman, C. (1979). Sex role stereotypes and justice for women. Qnmoano Delinquencx 25. 87-94 Feinman. C. (1980). Womenimhecriminauustmmm New York: Praeger Publishers. F IshbeIn M. & Ajzen, I. (1974). Attitudes towards objects as predictors of single and multiple behavior criteria. PsmholooioaLMow, 81, 59-74. Fishbein. M. &Aizen l (1975) Window Wren Reading. MA Addison-Wesley Fishbein, M. ,,&Ajzen l. (1981). Acceptance, yielding, and impact: Cognitive processes in persuasion. In R. E. Petty, T..M Ostrom, &T.C. Brock (Eds) WWW Hillsdale. NJ: Erlbaum 217 Fox, JG. (1984). Women's prison policy, prisoner activism, and the impact of the contemporary feminist movement: A case study. W. M, 15-36. Freedman, E. (1981). -' ‘ -- ~ ' 13mm. Ann Arbor. Universityof Michigan Press. Gagne, PL. (1992). Appalachian women: Wolence and social control. Journal Wally. 23. 387-415. Genders, E. 8. Player, 8. (1987). Women In prison. The treatment, the control and the experience. In P. Carlen & A Worrall, Conoononmajnoiuatioo, pp. 161 1-76. Philadelphia: Open University Press. Green, E, Hebron, S. & Woodward, D. (1987). Women, leisure and social control. In Hanmer, J. 8. Maynard, M. (Eds.,) Wm oontrol, pp. 75-92. Houndmills, Basingstoke, Hampshire, England. Macmillan Press. Hahn, N.F. (1980). Too dumb to know better". Cacogenic family studies and the criminology of women. Criminology, 18, 3-25. Hamilton, D. (1979). A cognitive-attributional analysis of stereotyping. W. 12. 53-84 Hamilton 0!. (Ed) (1981) WWW intorgrouomnayior, pp. 333-353. Hillsdale, NJ: Lawrence Erlbaum Associates. Harry, J. (1995). Sports ideology, attitudes toward women, and anti- homosexual attitudes. Soxfiolos, 32, 109-116. Heidensohn, F. (1985). Womananogjma. Houndmills, Great Britain: MacMiIIan. Henriques, Z.W. (1982). ‘ f ~ ' ‘ . . - angLanalytioaLstuoy. Lanham, MD: University Press ofAmerica, Inc. Herbert, R. (1985). Women's prisons: An equal protection evaluation. Ina IaleLanLEumal. %. 1182-1206. Houser, B.B., & Beckman, L.J. ( 1980). Background characteristics and women's dual-role attitudes. SaxBolao, 5, 355-366. 218 Huddy, L. (1994). The political significance of voters' gender stereotypes. W 4 169-193. Hutter, B. &Williams, G. (1980). Controlling women: The normal and the deviant. In. B. Hutter&G. Wlliams, (Eds.,) MW normaLanthedexiant pp 9-39 Innes, J. M. (1 993). Knowledge of gender stereotypes and attitudes toward women: A preliminary report. PsychologioaLBoporta,Z3, 1005-1006. Jackman, M. R. & Senter, M. S. (1980). Images of social groups: Categorical or qualified? W. 44. 341-361 Jackman, M. R. & Muha, M. J. (1984). Education and intergroup attitudes. Moral enlightenment, superficial democratic commitment, or ideological refinement? WM. 43. 751-769 Jackson, L.A., Sullivan, L.A., &Hodge. (1993). Cognition and affect In evaluation of stereotyped members. MmalofjmoiaLPsyonology, 123, 659-672. Janovw'tz, M. (1975). Sociological theoryand social control. W oLSmoIoov. 31. 82-108 Jones, G.P., & Jacklin, ON. (1988). Changes in sexist attitudes toward women during Introductory Women's and Men's Studies courses. Saxfiolos, 13, 61 1-622. Jones, M. (1991). Stereotyping hispanics and whites. perceived differences In social roles as a determinant of ethnic stereotypes. JoumaLoLSooial 8313091991. 1.3.1. 469-476. Kim. J., &Mueller, CW. (1978). WWW timIoDolt. Newbury Park, CA: Sage Publications, Inc. Klein, D. ( 1982) The etiology of female crime: A review of the literature. In B R Price and N J Sokoloff (Ede). W W. pp 35-60 New York Clark Boardman Company, Ltd. Lalonde, R. N. & Gardner, R. C. (1989). An intergroup perspective on stereotype organization and processing. WM. 23. 289-303. 219 LaPiere, R..T (1975). Attitudesv. actions. lnA..E Liska, (Ed), InoConoiatonoy York: Academic Press. LeFlore, L. 8. Holston, M. A. (1989). Perceived importance of parenting behaviors as reported by inmate mothers: An exploratory study. Journal Lengennann, P. M. 8. Wallace, RA. (1985). Theoretical overview. Social control. In P. M. Lengennann 8. R. A. Wallace (Eds.,) GonoorjnAmorioan; Wage pp 193? Englewood Cliffs N-JJ Prentice-Hall, Inc. Lippman, W. (1949). Puolioninjon. New York: The MacMillan Company. Liska, A. E. (1974). Emergent issues on the attitude-behavior controversy. W. 3.3 261-272 LlSka. AE (1975) (Ed) W ImpactoLAttituslurLBehavior New York: Academic Press LomaX. RG (1992) WWW BahavioraLSgienEs. New York Longman Lundberg, D., Sheckley, A., 8. Vuelkar, T. (1975). An exploration of the feelings and attitudes of women separated from their children due to incarceration. Masters Thesis. Portland, OR: Portland State University. Mahan, S. (1982). unflLMotIJora. Palo Alto, CA R&E Research Associates. Maher, L. (1992). Punishment and welfare: Crack cocaine and the regulation of mothering. HomemansmflminaLJustice. 3. 35-70. Mann, CR. (1984). Pomalomrrmnooolinguonoy. The University ofAlabama Press. McAIIister, I. 8 Moore, R. (1991). Social distance among Australian ethnic droops WW1; 95-100. McCarthy, B. (1980) Inmate mothers The problems of separation and. reintegration. _ , - _. ' 4,199-212. 220 McKinney, K (1987). Age and gender differences in college students' attitudes toward women: A replication and extension. Soxfioloa, 11, 353-358. Morash, M, Haarr, R. N., & Rucker, L. (1994). Acomparison of programming for women and men in U. S. prisons in the 1980s. Crimoarfimlmouonoy, 40,197-221. Morrison, T.G., Bell, E.M., Morrison, M.A., Murray, C.A., & O'Connor, W. (1994). An examination of adolescents’ salary expectations and gender-based occupational stereotyping. W. 25, 178-193. Moyer, I. L. (1984). Deceptions and realities of life In women's prisons. Ina W. 54. 45-56 Muir, D. E. (1991). 'White" fraternity and sorority attitudes toward "Blacks" on a deep-South campus W. 11 93-103 Murphy, W.F., &Tanenhaus, J. (1972). Ino_$tuoy_of_Puolio_Laws. New York: Random House. Naffine, N. (1987). - =_ ' Sydney, Australia: Allen & Unwin. Neto, V. 8. Ranier, L. (1983). Mother and wife locked up: A day with the family. IbeEnsonJoumal.53.124-141 Netting, N. S. (1991). Chinese aloofness from other groups. Social distance data from a city in British Columbia. W, 15, 101-.104 Norusis, M.J. (1982). Woo. NewYork: McGraw- Hill. Norusis, M.J. (1990). W. Chicago: SPSS, Incorporated. Owen, C. A., Eisner, H. C. 8. McFauI, T. R. (1981). A half-century of social distance research: National replication of the Bogardus' studies. Wasatch. E, 80-98. Park, R. (1924). The concept of social distance. ,lournaLoLAppljoofimology. 3, 339-344. 221 Pass, M. G. (1981). Race relations and the implications of education within prison. - - - . 145-1 51. Pass, M. G. (1987). Prison inmates express less social distance from minorities than do college students We]. 11. 1987 Petty RE .&Caoi0pp0.JT (1981) Attitudesaaiflsrsuaslanzlaschaarl ContammraiyApproaonos. Dubuque, Iowa: Wm. C. Brown Company. Pollock-Byme, J.M. (1990). W. Pacific Grove, CA: Brooks/Cole Publishing Company. Pollock, J.M. (1984) Women wn'll be women: Correctional officers' perceptions of the emotionality of women inmates. Inoorisonjournal, M. 84-91. Pomeroy. 8.8. (1975). Waves. New York: Schocken Books. Pratto, F. & Bargh, J. A. (1991). Stereotyping based on appareme individuating information: Trait and global components of sex stereotypes under attention overload JouraaIeLExreameataLfiEiaLEsxcaoloox. 21. 26-42. Fries. B R &Sokoloff NJ (Ede) (1982) IaeeamraalJchsexstsmeart WOW New York: Clark Boardman Company, Ltd. Rafter. N. (1990). (2d Ed). W New Brunswick, NJ: Transaction Publishers. Rea LM &Parker RA (1992) W mmorononsiyoguldo. San Francisco: Jossey-Bass Publishers. Rickman, S. ( 1983) Stereotypes by Black college students revisited. Paporsjn theSmaLScieares. 3. 21-28 Ronis, D.L., Yates, J.F., & Kirscht, JP. (1982). Attitudes, decisions and habits as determinants of repeated behavior. In A.R. Pratkanis, S.J. Breckler, & AG. Greenwald. (Eds). Attfludsflarctureaaifiuacma pp- 213-239- Hillsdale, NJ: Lawrence Erlbaum Associates. 222 Rothbart, M, Fulero, 6., Jensen, C. Howard, J, 8 Burrell, P. (1978). From individual to group impressions: Availability heuristics In stereotype formation JoumaleLEmameataLSocialflsxcaolagx.14.237-255 St. Pierre, R., Herendeen, N.M., Moore, D.S. &Nagle,A.M (1994). Does occupational stereotyping still exist? .loumaloLPsyohology, 123, 589- 598. Sargent, J. P. (1984). The evolution of a stereotype. Paternalism and the female inmate. Ino_Pn_son_Joumal, 54, 37-44. Schroeder, K.A., Blood, LL, 8 Malusso, D. (1992). An intergenerational analysis of expectations for women's career and family roles. Soxfiolos, 25, 273-291. Schwartz, S. (1978). Temporal instability as a moderator of the attitude- behavior relationship JorrraaLoLEsrseaalihraarlfioeiaLEsyaaalaov. E. 715-724. Schweber, C. 8 Feinman, C. (1985). The impact of legally mandated change on women prisoners CrimiaauusticeEaIIIIEaarIflQmea. 13. 4 Sherman, S. 8 Fazio, R. (1983). Parallels between attitudes and traits as predictors of behavior. ,lournaLoLPorsonality, 51, 309-345. Schuman, H. 8 Johnson, M. P. (1976). Attitudes and behavior. In A Inkeles; J. Coleman, 8 N. Smelser, (Eds), Annualfioyiowrotjooiology, Vol 2, pp. 161-207. Pala Alto, California. Sigelman, C. K, Sigelman, L. Walkosz, B.J., 8Nitz, M (1995). Black candidates, white voters. Understanding racial bias In political perceptions AmsaEaJouraaLoLBQlitiaalSeieace. 33. 243-265. Sills 01-. (Ed)- (1968). latsmatiaaaljarxclaaediamaeMaLSsisaass. 11. New York: Macmillan. Simon. R.J. &LandiS. J. (1991). W W. Lexington, MA: Lexington Books. Smart. C. 8- Smart. 8. (1978). flomsaeeAHaIIILaadsosiaLcaatral London: Routledge 8 Kegan Paul. 223 Smart. C. (1977). lALomanameaaoeamIaolooijomramlouo. London: Routledge 8 Kegan Paul. Smith, BA. (1990). The female prisoner in Ireland, 1855-1878. F_e_doral Probation. 3. 69-81. Sparrow, KH. 8 Chretien, D. M (1993). Social distance perceptions of racial and ethnic groups by college students: A research note. Sooiologioal Soootarra.13. 277-288 Spence, J.T., 8 Helmreich, R.L. (1979). Comparison of masculine and feminine personality attributes and sex-role attitudes across age groups. Doxoloomoatalfisxoholoox. 15. 583-584. Street, 8., Kimmel, E.B., 8 Kromrey, JD. (1995). Revisiting university student gender role perceptions. Soxfiolos, E, 193-201. Szymanski, LA, Devlin, A.S., Chrisler, JO, 8 Vyse, SA (1993). Gender role and attitudes towards rape in male and female college students. so); Bolao, 29, 37-57. Taylor, S. E. (1981). A categorization approach to stereotyping. In D. L. Hamilton. (Ed..) Cooaiflxeorooossosjaeomtxoiaoeaojateiorouo oohayior, pp. 83-114. Hillsdale, NJ: Lawrence Erlbaum Associates. Weigel, R. H. 8 Newman, LS. (1976). Increasing attitude-behavior correspondence by broadening the scope of the behavioral measure. JoumaLoLEorsoaalItLaaojEiaLEsxoaologx. 33. 793-802 Weisheit, R. (1985). Trends in programs for female offenders: The use of private agencies as service providers. lntomatjonalJoumaLoLfoonoor Wm. 23. 35-42 Welter, B. (1973). The cultoftrue womanhood. 1820-1860. InJ. E. Friedman &WG Shade (EdS) QuLAmoarerLsIstoLsJomoomAmoaoaaafiaao thought, pp. 96-123. Boston: Allyn and Bacon. West, C. 8 Zimmerman, D. H. (1991). Doing gender. In J. Lorber and Farrell, S--.A (Ede). IaoEoiaLEastantioaoLooaoor. pp 7- 12 Newbury Park. CA Sage Publications. 224 Wicker, A. W. (1969). Attitudes versus actions. The relationship of verbal and overt behavioral responses to attitude objects. W. 25, 41 -78. Wilder, D. A (1981). Perceiving persons as a group: Categorization and intergroup relations. In D. L. Hamilton (Ed), CognitiyaProoaasoajn StereotyolaoeaoJatoroLoooBeaavior. pp 213-257 Hillsdale NJ Lawrence Erlbaum Associates. Williams, F.M8McShane, MD. (1988). Criminolmioallhaory. Englewood Cliffs, NJ: Prentice Hall. Young, V.D. ( 1986). Gender expectation and their impact on Black female offenders and victims. Justmuartorly, 3, 305-327. Zupan, L. L. (1992). Men guarding women. An analysis of the employment of male correction officers In prisons for women. ,loumaLoLCriminaLJuotioo, 29297-309. "IIIIIIIIIIIIIIIIIIIIIIS