fit ‘ T ,. . ‘ fin,“- tgfitilzl‘r v :1! r". .. W). ct games; “..>. Inn’.‘ . . ll: “NJ.” vol- .. 't 0 it on“! .. kHN-v ~ {1.1. .7 r i: ll I‘M] l. .L ‘ . sifF‘; ! J r #3 .‘ _ w z ...x»; a . xuuv 3 ! 9‘ ' ‘r. ,. :". :l‘lbisg iii-mus. .I "i l o 2;} v ‘ . ‘uvstu . (.99‘111‘ Ill! rs 1‘0.‘cllv!.0tp .531. I: JEIIOt‘lvbiiyv’t’uR‘1ulP‘n . (Y3.1|.i!$ I’lilttro. .lb.br.!.ivl....ncl - ~ .I V I} u u - “'II..J>'I1"V|‘.'.V§" volion vlr‘l: In-.. -'1.r' H‘ r’IV .I. h. ‘v e l... .'.|_ - . 1% ".I v.) . It. I: ., kins ’I...IIP(O0 Iv] .vbAt In! ~lll|lr| . e L, .|;v.. ‘ “I“: ‘ . ‘votnvvn>.t4;c..¥s..x .vtx.§sivn‘.uol.v.- ....;.o.l..!nf!.- ‘9’}: )Ila :7 P73: .6. It‘lvhlcnou‘ :II‘... V. LJVtQIl.‘€ 351921 ‘3' filillé I JV..- . r '3‘:- . II. 323...)! in“ :- h r hiiznvv in? .s.i..s ‘ 3-2.1331»... 5 ...l.... ....Iit.y..,- - 1].]. , . 3.1.1.?! {pupil}? :11 ..£.a&..iirl.5-!l.ti in". 1 II. www.3su..u.ual.flv.;n 3L... w I fo.l I‘D '.v».v’\‘.“r"n‘1 tn . . «an NIVERSITY LIBRARIES IIIIIIIIIII IIIIIIIIIIIIIIIIIIIII I II I IIIIII 312 29300 This is to certify that the dissertation entitled The Relationship of Family Environment, Mother's Influence, and Pregnant Adolescents' Decisions to Keep or Release Their Babies for Adoption presented by Susan Neda Popovich has been accepted towards fulfillment of the requirements for /7./7 [3 degreein %7(/;“/‘75 I /——-" /M ét;:’:;€& Date May 25, 1990 MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771 ___.._——.——_.._ —— LIBRARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE We 6 193? __——— ,,,-5W THE RELATIONSHIP OF FAMILY ENVIRONMENT, MOTHER’S INFLUENCE, ; ’ AND PREGNANT ADOLESCENTS' DECISIONS TO '4 1 KEEP OR RELEASE THEIR BABIES FOR ADOPTION - ‘ I BY Susan Neda Popovich '2 I A DISSERTATION cflh‘ 4 Submitted to Michigan State University 14+ in partial fulfillment of the requirements for the degree'of . sift-‘1 DOCTOR OF PHILOSOPHY - 1990 Department of Psychology ”13": . "recrsicnr-’ abormr " 64-7r533'3 ABSTRACT THE RELATIONSHIP OF FAMILY ENVIRONMENT, MOTHER’S INFLUENCE, AND PREGNANT ADOLESCENTS’ DECISIONS TO KEEP OR RELEASE THEIR BABIES FOR ADOPTION BY Susan Neda Popovich Adolescent pregnancy is a widely recognized social problem. m W IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII “ Not much is known about why adolescents do not choose adoption. This study examined demographic characteristics and attitudes of adolescent “keepers" and "releasers" and the impact of family environment and the adolescent’s mother’s influence on the decision to keep or release. Data for this study were drawn from a larger research project on adoption decisions. Adolescents who kept their children were subtyped according to whether they considered no other alternative (n=14), also considered abortion (n=14), or also considered adoption (n=12), and these keeper subtypes were compared to a group of adolescents who released their children for adoption (n=14). Respondents were recruited at hospitals following the babies’ births and through adoption agencies. Data were collected by individual interviews. All respondents were white, never-married, and had no older children at home. Results showed that demographically, adolescents who chose to keep were very similar to adolescents who chose adoption. However, "releasers" were further along when they realized the pregnancy, reported reacting more negatively, and had more specific educational plans compared to adolescents who kept. Attitudes toward adoption differed among groups, with releasers less concerned about how a child would be integrated into an adoptive family compared to adolescents who did not consider any alternative besides keeping. Most of the adolescents’ mothers reacted negatively to the pregnancy, but offered their daughters emotional support and practical help. Adolescents reported that mothers did not influence their decisions strongly, but mother’s preference for outcome was related to the outcome decision. Family environment could not be reliably measured in this sample, and its impact on pregnancy resolution decisions among adolescents remains an area for future research. Findings support a social-structural theory of adolescent pregnancy. Characteristics of adolescents most likely to consider adoption are discussed. This dissertation is dedicated to my parents, whose support for my educational endeavors has been unconditional. ACKNOWLEDGMENTS I would like to extend my thanks to: David Kallen, for your support in this project from the beginning, for your seemingly endless availability, and for your patience in letting me set my own pace. Bob Caldwell, for your willingness to co-chair my committee. My experience was enriched by the high standards you set forth, your attention to detail, and your support. Ellen Strommen, for your thought-provoking contributions to this work. Your presence in a male-dominated department and on my committee provided me with a welcome role model. Gary Stollak, for your involvement in this dissertation and other projects over the years. Ginny, Karen, Carol, Kathy, Betty, Tim, Val, Carita, and my sister Carol, for believing that I could do this and reminding me that I could when I most needed to hear it. TABLE OF CONTENTS List of Tables ....................................... x List of Figures .... .................................. xi Chapter One: Adolescent Pregnancy: The Problem ..... 1 Adolescent Pregnancy and Fertility ................. 1 Pregnancy Statistics ............................. 1 Fertility Statistics ..... ........................ 2 Adolescent Population .. ............. . ............ 3 Adolescent Sexual Behavior ....................... 4 Changing Trends in Pregnancy Outcome ............. 5 Consequences of Teenage Parenthood ................. 10 Health Risks ..................................... 11 Family Structure and Size ........................ 12 Educational Attainment ........................... 15 Economic Status .... .............................. 15 Long Term Follow—Up .............................. 16 Societal Costs ........ ...... ..................... 17 Definition of the Problem .. ............... . ........ 17 Adoption as a Possible Solution ............. . ...... 18 Purpose of This Research ........... .. .............. 20 Chapter Two: Review of the Literature: Adolescent Pregnancy Resolution Decision Making ... 22 Aborters vs. Keepers ..... . ......................... 23 Demographic Variables ... ......................... 23 Family Variables .................... . ............ 25 Other Relationships ..... ......................... 26 Releasers vs. Keepers ............. . ................ 26 Factors Related to Adoption ...................... 27 Changing Perspectives ......................... ... 34 Adolescent Pregnancy Resolution Decision Making .... 37 Decision Making Models ........................... 37 Reasons for Not Releasing ............... ......... 42 Mother’s Influence .........V .............. . ....... 44 Purpose of This Research ...................... ..... 48 Hypotheses ............... .......... ................ 51 Hypothesis #1 ....................... . ............ 51 Hypothesis #2 .................................... 51 Hypothesis #3 ......... ....... ........... . ........ 52 Hypothesis #4 ....... . ............................ 52 Hypothesis #5 .................................... 52 Hypothesis #6 .................................... 52 Hypothesis #7 ...... ...... ........................ 53 Hypothesis #8 .................................... 53 Hypothesis #9 ....................... . ............ 54 Hypothesis #10 ................. . ................. 54 Chapter Three: Methodology ..... ..................... 55 Overall Study ...................................... 55 Design .......... ................................. 55 Data Collection Procedures ....................... 55 Keepers ............ ... ............ . ............ 55 Releasers ............ .. ........................ 57 Present Study ................................... ... 58 Sample .............. . ........ . ................... 58 Releasers ...................................... 58 Keepers ........................................ 58 Variables and Measurement ........................ 59 Individual Variables ........................... ' 60 Family Variables ............................... 63 Pregnancy-Related Variables ............. . ...... 67 Chapter Four: Results .............. . ................ 71 Sample Description ................................. 71 Demographic Data ................................. 71 Family Background ............... . ................ 72 Pregnancy Variables ...................... . ....... 75 Analysis of Hypotheses .. ........................... 78 Hypothesis #1 .................................... 78 Hypothesis #2 ........................... .... ..... 78 Hypothesis #3 . ................................... 79 Hypothesis #4 .............. . ..................... _81 Hypothesis #5 .................................... 81 Hypothesis #6 .................................... 82 viii Hypothesis #7 .................................... 83 Hypothesis #8 ............. . ...................... 84 Hypothesis #9 ............... . .................... 87 Hypothesis #10 ..... .............................. 89 Summary of Results .................... . ............ 90 Chapter Five: Discussion ...................... . ..... 94 Releaser Characteristics ... ........................ 94 Mother’s Influence ... ..... ...... ................... 98 Interpretation of Differences ...................... 100 Family Environment ............... .. ................ 102 A Social—Structural View of Adolescent Pregnancy ... 104 Is Adoption 3 Solution? ............................ 106 Appendix A ...... ........ . ........ . ................... 109 Appendix B ..... .... .......... ... .............. .. ..... 115 Individual Variables .......... .................... . 115 Family Variables . ....... . .......................... 119 Pregnancy—Related Variables ........................ 124 Appendix C ........................................... 131 Step One ........................................... 131 Step Two . ...... . ................................... 132 Step Three ......................................... 133 Step Four ........ .................................. 133 Step Five ..... ............ . ........................ 134 Conclusions ........................................ 134 References ..... . ..... . ............................... 136 Table Table Table Table Table Table Table Table Table Table Table Table Table Table _L LIST OF TABLES Adolescent Fertility: 1955—1983 ........... Description of Sample ...................... Family Environment Subscale Descriptions Summary of Variables ....................... Religious Affiliation by Final Decision Mother’s Employment Status at Time of Pregnancy ............................... Timing of Pregnancy Realization by Decision Outcome .................................... Adolescents’ Reactions by Outcome Group Abortion Attitudes by Outcome Group ........ Relationship between Outcome and Educational Plans ..... . .................... ANOVA of "Afraid” by Outcome Group ......... Perception of Own Influence on Decision by Group ................................... Mother’s Outcome Preference by Outcome Group ... ........................... Reliabilities for the FES (Form S) Subscales ........... .. .. ............. ..... 60 66 69 73 75 76 77 79 80 84 85 88 132 Figure 1 LIST OF FIGURES Application of Pregnancy Decision Making Model (Eisen et al., 1983) to This Study 50 CHAPTER ONE ADOLESCENT PREGNANCY: THE PROBLEM Adolescent pregnancy is a widely recognized problem in this country. Public concern is evidenced by the prevalence of teenage parenthood as a topic in both the professional literature and the popular media. Adolescent pregnancy is a complex phenomenon, related to political, moral, and policy issues which can affect the physical and psychological well being of both the adolescent mother and her child. Adoption has been posited as one solution to this problem, yet is infrequently chosen by adolescent mothers. This study examines the influence of the adolescent’s mother and family environment on her decision to keep or release her baby. Adolescent Pregnancy and Fertility E i! I' !' Each year over one million adolescents in this country get pregnant (Furstenburg & Brooks-Gunn, 1985; National Research Council, 1987). This is approximately one out of every ten teenage girls. Forty percent of the current population of fifteen—year-old girls are expected to conceive while they are still teenagers (Mecklenburg & Thompson, 1983). The number of adolescent pregnancies has increased during the 19703, and black adolescents experience a higher incidence of unintended pregnancies compared to 2 whites (Ascher, 1985; National Research Council, 1987). The majority of teenage pregnancies are unintended (Alan Guttmacher Institute, 1981; Zelnick & Kantner, 1980), and it is further estimated that three quarters of all the pregnancies in the 15- to 19-year-old age group occur to unmarried adolescents (O’Connell & Rogers, 1984). However, not all of these pregnancies result in a birth. Depending on the source, from one-third (Olson, 1980; Phipps-Yonas, 1980) to 40% (National Research Council, 1987) of women under age 20 or one-half of the young women under the age of 18 (Henshaw, Binkin, Blaine, & Smith, 1985) voluntarily terminate their pregnancies by abortion. Statistics show that about 13% of the pregnancies among adolescents under age 20 end in miscarriage (National Research Council, 1987). Thus, close to one-half of all adolescent pregnancies result in birth (National Research Council, 1987; Olson, 1980), and an estimated 90% to 95% of these adolescent mothers keep and raise their babies (Ascher, 1985; Bachrach, 1986; Furstenberg & Brooks-Gunn, 1985). E ! 1'! E! l' !' Although public attention is captured by the "alarming" number of adolescent pregnancies each year, adolescent fertility is actually decreasing (Chilman, 1979; Furstenberg & Brooks-Gunn, 1985). There was a 20% drop in the actual number of births to adolescents from 1970 to 1984, with the A 3 greatest decreases evident among 18— and 19-year—olds as compared to younger teens. Births among both black and white teens were fewer during this period, with greater decreases among older whites and younger blacks (National Research Council, 1987). The apparent contradiction between the growing number of adolescent pregnancies and declining fertility rates can be explained when the many variables that influence adolescent pregnancy and childbearing are examined. Among these variables are the size of the adolescent population, the incidence of adolescent sexual activity, and _ m II 1 ! E 1 !' Adolescent pregnancy and fertility statistics are directly dependent on the overall size of the adolescent population, and it has been changing over the last couple of decades. The size of the adolescent population was a function of the post-World War II “baby boom", peaking in 1976 at 21.4 million (Chilman, 1978, 1979; National Research Council, 1987). Since then, the number of adolescents aged 15 to 19 has been steadily declining, although it continues to range between 7% and 10% of the total U.S. population (National Research Council, 1987). Not surprisingly, in the 1960s and 19703 when the overall adolescent population was larger, there were more teenagers at risk for pregnancy and more births were reported; this increase in number of 4 teenage births was without doubt one factor which aroused public concern. However, when the birthrates among this age group during the same years are examined, a decrease is evident. These figures are shown in Table 1. Table 1 Adolescent Fertility: 1955-1983 Age 1955 1960 1970 1980 1983 Number of Births (in thousands) 15-19 Total 484 587 645 552 489 White 373 459 464 388 338 Nonwhite 111 129 172 150 137 Birthrates (per 1,000 women) 15-19 Total 90.3 89.1 68.3 53.0 51.7 White 79.1 79.4 57.4 44.7 43.6 Nonwhite 167.2 156.1 147.7 100.0 95.5 Ngtg. From "Teenage childbearing: Causes, consequences, and remedies" by F. Furstenberg and J. Brooks- -Gunn. In L. H. Aiken and D. Mechanic (Eds. ) Applications of social science 59 gljnigal medicine and heal ED 991 jgy (Table 15.1, p. 309), 1986, New Brunswick, NJ: Rutgers University Press. Copyright 1986 by Rutgers, the State University of New Jersey. Reprinted with permission of Rutgers University Press. WW2: The actual number of adolescent pregnancies during the 19605 and 19705 increased also as a result of the "sexual revolution" (Chilman, 1979; Furstenberg & Brooks-Gunn, 1985; Kallen & Stephenson, 1982). The greater incidence of 5 premarital intercourse among adolescents put more adolescents at risk for pregnancy. Zelnik and Kantner (1980) reported that by 1979 nearly 50% of all adolescent girls were sexually active. Among urban teenage women, this represents an increase of two-thirds of the proportion who had engaged in intercourse during the 19705. It is also well established in the literature that significantly more black adolescents engage in sexual intercourse than whites, and become sexually active at a younger age (Furstenberg, Morgan, Moore, & Peterson, 1987), one factor that accounts for the higher incidence of unintended pregnancies among blacks. These racial differences, although still evident, are diminishing; during the 19705, sexual activity among white, urban teenage women doubled, while for blacks it only rose slightly (Zelnik & Kantner, 1980). Increases in sexual activity caused a rise in the pregnancy rate in the 19705, but not nearly of the same magnitude as the increase in sexual activity; this is attributable to increased contraceptive use (National Research Council, 1987). There was a rise in contraceptive use among single, sexually active teens aged 15 to 19 in the 19705, although the evidence still indicates that adolescents are generally poor contraceptors. Eighty—five percent of this group reported having ever used some type of contraception, but close to 40% indicated that they only 6 used contraception ”sometimes". Research also shows that white adolescents are better contraceptors than blacks, who are more sexually active and therefore doubly at risk for unplanned pregnancy (National Research Council, 1987). Ascher (1985) pointed out that studies examining racial differences in contraceptive use are generally confounded by failure to control for social class differences, and cautioned that conclusions about racial differences in contraceptive use are therefore unclear. H . I ! . E ; ! Adolescent birthrates have been greatly influenced by society’s changing attitudes toward alternative outcomes for pregnant adolescents. The availability of abortion and greater acceptance of single parenthood have impacted adolescent birthrates. Availability of these options has also affected the more traditional outcome of teen pregnancy, namely adoption. Anggtign. Changes in federal policy by the Supreme Court in 1973 made abortion a legal alternative for pregnancy resolution for adolescents (Chilman, 1978, 1979). The number of abortions obtained by 15- to 19-year-olds between 1971 and 1973 nearly doubled. However, after the initial escalation following legalization, the proportion of abortions performed among teenagers leveled off during the late 19705 and appears to be declining in the 19805 (National Research Council, 1987). This may be related to 7 withdrawal of federal funds for abortions, except in cases of rape, incest, or threats to the mother’s health (Chilman, 1979), since an unmet need for abortion is reported to exist, especially among black adolescents (Alan Guttmacher Institute, 1981). Age and racial differences in abortion rates have been documented. While one estimate of the number of pregnancies that are aborted among 15- to 19-year-olds is 40%, for teens younger than 15 years, the ratio of abortions to live births is nearly 1.4 to 1 (National Research Council, 1987). Pregnant adolescents who are black have been traditionally less likely than those who are white to terminate an unintended pregnancy by abortion (Eisen, Zellman, Leibowitz, Chow, & Evans, 1983; National Research Council, 1987). More recent data, however, suggest that these differences may be diminishing (Henshaw & Van Vort, 1989). However, abortion rates, which reflect the number of abortions per 1,000 adolescents, are higher among blacks compared to white adolescents. In 1980 and 1981, black teens in the 15- to 19-year-old age group had an abortion rate about twice as high as that for whites (Ascher, 1985; National Research Council, 1987). For black teens under 15, the rates were 5 times the abortion rates among white teens in the same age group. Abortion rates among blacks are higher than among whites because of the higher incidence of sexual activity and unplanned adolescent pregnancies among blacks. 8 51nglg_nanentnggg. Over one-half of all births to adolescents occur to single adolescents (Furstenberg & Brooks-Gunn, 1985; National Research Council, 1987). This represents a change over recent decades, when one means of resolving teenage pregnancy was by marriage. Black teens are significantly more likely than whites to be unmarried mothers; in 1984, the rate of childbirth outside marriage among blacks aged 15 to 19 was nearly 5 times that among whites in the same age group (National Research Council, 1987). Black teens at all ages are more likely than whites to be unmarried (Eisen et al., 1983; Moore, Simms, & Betsey, 1986). Although the highest proportion of marriages to legitimate a birth occurs among older, white teenagers (Ascher, 1985; Chilman, 1980; Furstenberg & Brooks-Gunn, 1985; National Research Council, 1987), the overall increase in births to unmarried teenagers is accounted for by increases in the number of births to white, unmarried adolescents (National Research Council, 1987). The growth of childbearing outside marriage reflects both the increase in average age for first marriage for the population as a whole, as well as changing societal attitudes about the necessity of marriage to legitimate birth (Eisen et al., 1983; National Research Council, 1987). Adolescents constitute a smaller proportion of unmarried parents; single childbearing outside marriage has also become more common among adult women over the last 15 years. 9 Adoption. In spite of the fact that most teenage pregnancies are unintended, relatively few babies are released for adoption. Current statistics about adoption are difficult to estimate because there have not been systematically collected national data on adoption since 1974 (Bachrach, 1983). Agency estimates from select states that are available tend to underrepresent adoption rates since they do not include statistics from private adoptions (National Research Council, 1987). It is evident, however, that adoption rates are declining. National estimates for the period 1957 to 1970 indicate that adoptions rose from 90,000 in 1957 to 175,000 in 1970, dropping to 149,000 in 1974. National data thereafter are nonexistent. It should be noted that these figures are not limited to children of adolescent mothers (Bachrach, 1983). In 1970, an estimated 90% of pregnant, unwed adolescents relinquished their babies for adoption, but by 1978, 90% of young mothers were choosing to keep their children (Phipps-Yonas, 1980; Resnick, 1984). There has been little change in adolescent mothers’ inclinations to relinquish their babies since the mid-1970’s (Bachrach, 1986). For example, agency data more recently collected in Minnesota indicated that although there were 6,107 births to teenagers in 1982, only 45 newborns were released for adoption throughout 1983 and 1984 (Children’s Home Society of Minnesota, cited in National Research Council, 1987). 10 The change in adoption trends represents differences primarily in the behavior of white adolescents. — m ”(Furstenberg 8. Brooks-Gunn, 1985; Phipps-Yonas, 1980; Resnick, 1984). Reported estimates in 1971 indicate that 72% of white and 92% of black adolescent mothers aged 15 to 19 kept their infants, compared to 1982 estimates, which showed that 91% of whites and 95% of blacks chose to parent (National Research Council, 1987). ‘ (Bachrach, 1986; Furstenberg & Brooks-Gunn, 1985). Clearly, the tendency of single adolescent mothers to relinquish their babies for adoption has decreased since 1970. Researchers differ in the emphasis given to each factor, but the decline in adoption rates among adolescents is attributable to the availability of legalized abortion and more tolerant societal attitudes toward single parenthood (Resnick, 1984). Consequences of Teenage Parenthood There is a vast literature on the consequences of early parenthood for teenage mothers and their children. Early childbearing has generally negative effects for the adolescent, her child, and society. These outcomes are significant in understanding why adolescent pregnancy and parenthood are considered problematic. M Adolescent pregnancy and childbirth are associated with higher health risks for both mother and child. Higher incidence of pregnancy and birth complications, maternal mortality, miscarriage, stillbirths, premature and low birthweight infants, and child abuse are reported among teens and their children compared to older women (Ascher, 1985; Mecklenburg & Thompson, 1983; National Research Council, 1987; Phipps-Yonas, 1980). This is especially true for teens aged 15 or younger (Phipps-Yonas, 1980); this group is twice as likely to have a premature or low birthweight baby and maternal death is 2 1/2 times as frequent, compared to mothers aged 20 to 24 years (Alan Guttmacher Institute, 1981; National Research Council, 1987). Early studies about the health risks associated with adolescent pregnancy confounded age effects with other variables such as socioeconomic status and race. However, an age effect is still evident when these other variables are controlled (National Research Council, 1987). Many of these negative effects of teen pregnancy and childbirth are attributable to poor prenatal care and nutrition (National Research council, 1987; Phipps-Yonas, 1980). Teens are notorious for poor eating habits, and even after the pregnancy is discovered, a proper diet is not always followed. In addition, many pregnant adolescents receive prenatal care only late in the pregnancy, if at all 12 (Alan Guttmacher Institute, 1981; Phipps-Yonas, 1980). Poor adolescents and their children are at highest risk for health complications and longer term developmental problems since they are less likely to have adequate prenatal care. The mental health of the adolescent mother who chooses to parent is also affected. Adolescents who keep and parent their children are under increased stress. Poor self-esteem and higher incidence of depression are found among parenting teens (Barth, Schinke, & Maxwell, 1983). E .1 E! I | 5' Although teens today are less likely than previous cohorts to marry in order to legitimate the birth, about one-half of the adolescents who give birth are married (O’Connell & Rogers, 1984). Older teenagers are more likely to marry to legitimate the birth compared to younger teens. The value of marriage among teenagers is questionable, because the marriages appear to be very unstable (Chilman, 1979). Compared to divorce rates among women who wait to marry in their 205, divorce rates are 3 times as high among young women who get married while still in their teens (Alan Guttmacher Institute, 1981; Furstenberg & Brooks-Gunn, 1985). For teens who do marry, the outcome is more favorable if the marriage occurs before the baby’s birth. Not only is the couple more likely to stay together compared to couples who marry after the birth, but the infants show more positive outcomes as well, such as higher birthweights 13 (McLaughlin, Grady, Billy, Landale, & Winges, 1986). Follow-up studies of unmarried teens who keep their babies indicate that the majority return with their children to live in their parental home, at least on a temporary basis; estimates range from 72% to 90% (Ascher, 1985; Clapp & Raab, 1978; Furstenberg & Brooks-Gunn, 1985; Furstenberg & Crawford, 1978; Mecklenburg & Thompson, 1983; Young, Berkman, & Rehr, 1978). It appears that single adolescent mothers are often disappointed by the emotional and financial assistance provided by the babies’ fathers, although they do attempt to maintain some involvement (Ascher, 1985), and therefore need to resort to family support if it is available. Burton’s work with black adolescent mothers and their mothers and grandmothers (cited in Elder, Caspi, & Burton, 1988) showed that the adolescent mothers’ perceptions of future life plans were related to their mothers’ reactions to the role of grandmother. Elder et al. (1988) pointed out that from a developmental perspective, adolescent parenthood can precipitate premature role transitions in the family across generations. Most young grandmothers rejected their premature roles as grandparent and were not active in caretaking of the child; in these cases, over half of the adolescents shifted parental responsibilities for the child to their own grandmothers. The remainder had little choice but to assume the primary caretaking role. The adolescents 14 who were able to shift parental responsibility to family members had future aspirations that included marriage, having more children, and owning a home. In contrast, parenting adolescents were focussed on finishing high school, and had little hope of marriage or career plans. Finally, having a child as a teen is correlated with future family size and spacing of births. Women who bear children as adolescents tend to have more children, more unintended births, more nonmarital births, and closer spacing of births compared to those who delay childbearing (Ascher, 1985; Chilman, 1979; National Research Council, 1987; Phipps-Yonas, 1980). For example, 60% of those who have additional children do so within two years of their first birth, with just under 20% becoming pregnant again within the following year (Ascher, 1985). Larger families are not planned, but these women are at greater risk for repeated pregnancies given their early start (Furstenberg & Brooks-Gunn, 1985). Early childbearers tend to have larger families by about one child, on the average (National Research Council, 1987); this reflects a diminishing difference in family size between early and later childbearers compared to previous decades and is possibly due to the increased availability of contraception and abortion (Furstenberg & Brooks-Gunn, 1985). El . ] El! . ! Early childbearing interrupts the expected developmental sequence of educational attainment. Even controlling for socioeconomic status, academic ability, and motivation, multiple studies provide empirical evidence to confirm that young women who give birth as adolescents have an average of fewer years in school and are less likely to complete high school or college compared to women who have first children at older ages (Alan Guttmacher Institute, 1981; National Research Council, 1987). Teens who become mothers before the age of 18 are more likely than older mothers to drop out of school, either at the time of birth or during the year preceding the birth (Ascher, 1985; National Research Council, 1987). Furthermore, an estimated 85% of the pregnant teens who drop out never return to school (Dunkle, 1984, cited in Ascher, 1985), nor are these young women likely to ever catch up or complete their educations (National Research Council, 1987). Ma Women who begin motherhood early are at greater risk for poverty. Given larger family sizes and educational deficits, it is not surprising that adolescent mothers have limited employment opportunities and a greater reliance on public assistance when compared to women who delay first births (National Research Council, 1987). Not only are these young women less likely to earn high wages, but they 16 are typically the economic heads of their households, since their marital relationships are unstable. The Alan Guttmacher Institute (1981) reported that young single mothers are seven times as likely to be poor as other women their age. Needless to say, the consequences of living in poverty are detrimental to the development of both mother and child. W Furstenberg, Brooks—Gunn, and Morgan (1987) reported a longitudinal study of adolescent mothers which gives a different perspective on the consequences of teen parenthood. Data from interviews conducted 17 years after the adolescents gave birth suggest that previous research findings exaggerate the plight of the adolescent mother in later life. Furstenberg and his colleagues reported that the majority of adolescent mothers in their sample managed to complete high school, to get off public assistance and find employment, and to control family size. Compared to the earlier 5-year follow—up in this study, the women were more successful except with respect to marital status; most of the mothers were single, separated, or divorced. Furstenberg et al. (1987) concluded, however, that although the stereotype of the adolescent mother in later life is exaggerated, it is in fact not completely incorrect. Many mothers may manage to break out of the poverty cycle by completing their educations, finding employment, and 17 managing family size, but they do not make out as well as they would have had they postponed parenthood. §Q§i§&§l_QQ§1§ For society, the cost of adolescent pregnancy is two- fold. First is the loss of potential contributions these young women may have otherwise made. Secondly, there are direct costs stemming from their economic reliance on public assistance. For 1975, an estimated 8.6 billion dollars in Aid to Families with Dependent Children (AFDC) was distributed to adolescent mothers (Ascher, 1985); this represented over 50% of the AFDC budget during that year. More recent estimates suggest these costs have doubled in the last decade (National Research Council, 1987). Definition of the Problem Adolescent pregnancy is a complex phenomenon and is considered problematic for a variety of reasons (National Research Council, 1987). For some, the problem is a moral issue and rests with the fact that adolescents are engaging in premarital, sexual intercourse at a young age. It is the sexual activity and not the pregnancy that is of primary concern from this perspective. Others, however, accept the changing societal trends in sexual behavior, contraceptive use, women’s roles, and family and work patterns, and do not expect adolescents to be exempt from these changes. Proponents of this view consider adolescent pregnancy problematic in that public policy does not sufficiently meet 18 the needs of these young women by helping them to avoid the personal and societal risks associated with unplanned conception. From yet another perspective, it is the outcome of the pregnancy that raises concern. Politically conservative groups have moral objections to the high number of abortions among teenagers, and identify the high rate of abortion as the main problem of adolescent pregnancy. Finally, others are concerned about the consequences of parenthood at so young an age, given that 95% of the pregnancies carried to term result in the adolescent mother keeping and raising her child. Adoption as a Possible Solution Given the many aspects of adolescent pregnancy that have been labeled as problematic, it is not surprising to find that multiple approaches have been taken to address the problem (National Research Council, 1987; Phipps—Yonas, 1980). For instance, primary prevention programs are aimed at delaying sexual activity among teens and increasing knowledge and availability of contraception. Interventions at secondary prevention levels also exist, including availability of abortion, and improved health care and parent training for pregnant teens. Finally, tertiary intervention programs emphasize improvement of the adolescent mother’s life situation through increased public assistance or continuing education and employment opportunities. -, 19 In the last few years, adoption has received more attention as a possible solution to adolescent pregnancy. This alternative to pregnancy resolution appeals to conservative groups who oppose abortion on moral grounds, and provides an alternative that eliminates some of the negative consequences of teen parenthood. In the past when adoption was more common, the single adolescent typically spent her pregnancy away from home, in a home for unwed mothers or with an out-of-town relative. The picture for adolescents who carry their pregnancies to term and choose to release their children has changed. Because of more social acceptance of pregnancy outside of marriage, adolescents who choose adoption over abortion or parenthood are no longer hidden away during their pregnancies (Musick, Handler, & Waddill, 1984; Phipps-Yonas, 1980). Furthermore, the adoption process itself is changing. In Michigan, for example, some agencies are offering the birth mother more choice in selecting the adoptive family with whom her infant will be placed. It is unclear how or if the changes in adoption process might affect adoption as an alternative for pregnancy resolution decisions. Renewed interest in adoption by policy makers is accompanied by a demand for adoptable infants. There is a trend toward delay in childbearing among married women. In 1979, over 10% of 30- to 34-year-old, married women in the U.S. were childless, and nearly one-half of them still 20 planned to have a child in the future (Bachrach, 1983). This trend may reflect the older average age for first marriage, as well as pursuit of careers outside the home. In any event, infertility is an issue for an increasing proportion of women (Bachrach, 1983). 80 according to current trends, the number of adoptive parents is growing while the current number of infants available for adoption is diminishing (Grow a Smith, 1971; Haring, 1975). Advocating adoption as a possible solution to adolescent pregnancy would make more infants available for adoption, without posing further problems of placement, given the present circumstances. Purpose of This Research Adoption appears to be one possible solution for adolescent pregnancy. However, not much is known about why adolescents do not elect this alternative more often. This study will compare a small number of adolescents who released their babies for adoption (“releasers") with adolescents who kept and parented their children ("keepers"). Because so many teenagers live at home at the time of the pregnancy and bring their children home (Clapp & Raab, 1978; Furstenberg & Crawford, 1978), this study will focus on family variables and maternal influences on the decision to keep or release the baby. Moore, Peterson, and Furstenberg (1986) pointed out that intervention may be more effective at the family level compared to community 21 services, since groups and persons closer to the adolescent have greater influence on her decisions about sexual activity, pregnancy, and pregnancy resolution. It is only when more knowledge about adolescent pregnancy resolution concerning adoption is acquired that policy decisions about the viability of adoption as a solution to adolescent pregnancy can be made. 1:.— CHAPTER TWO REVIEW OF THE LITERATURE: ADOLESCENT PREGNANCY RESOLUTION DECISION MAKING The research on adolescent pregnancy resolution can be divided into two general areas: (a) studies that describe demographic and psychosocial characteristics associated with outcome decision, and (b) studies that address the decision— making process, including the influence of significant others on the decision. The majority of studies are of the first type, and have been approached from both sociological and psychological perspectives. Adolescents who terminate their pregnancies by induced abortions (aborters) or who relinquish their babies for adoption (releasers) are typically compared to adolescents who carry their pregnancies to term and elect to parent (keepers). A majority of these studies deal with the decision about abortion, since the decision to surrender a baby for adoption is made so infrequently. This review will summarize the research findings pertaining to aborters, but will emphasize the adoption decision and family influences on the process. 22 23 Aborters vs. Keepers I I' H . I] B§§§_§ng_§E§. Black adolescents are less likely to seek abortion as a pregnancy resolution than are whites (Eisen et al., 1983; National Research Council, 1987; Olson, 1980; Zelnik, Kantner, & Ford, 1981). Social class differences also exist between aborters and keepers. Most investigators have shown that pregnant adolescents who choose abortion are from higher socioeconomic groups than are adolescents who keep their babies (Olson, 1980; Phipps— Yonas, 1980; Zelnik et al., 1981). Age. Age has been found to be a determinant of abortion. In her review of the literature on adolescents who choose abortion, Olson (1980) found that aborters are either in their very early teens (15 or younger) or later teens (18 and 19). Correspondingly, young women age 16 to 17 are more likely to keep and parent their children. Overall, though, the aborter is older than the keeper (Olson, 1980). Age is also related to timing of the abortion. The younger the adolescent, the longer the delay in obtaining an abortion (Alan Guttmacher, 1981). These delays have been explained as failure to recognize pregnancy symptoms (due to irregular menstrual cycles among younger adolescents) or as denial of the pregnancy. 24 Educatign. The adolescent who decides on abortion was usually doing well in school before the pregnancy (Eisen et al., 1983; Leibowitz, Eisen, & Chow, 1980), but not necessarily better than the average student (Olson, 1980). A number of studies reported that school performance distinguishes aborters from keepers, with keepers getting worse grades and having more academic difficulties (Evans, Selstad, & Welcher, 1976; Olson, 1980) or having completed less education (Evans, et al., 1976). Aborters also have been found to have higher educational and occupational goals than adolescents who decide to parent (Eisen et al., 1983; Leibowitz et al., 1980; Olson, 1980; Phipps-Yonas, 1980). Emnlgymgnt. Although aborters are just as likely as keepers to be supported economically by their families of origin at the time of pregnancy, aborters tend to be more financially independent from their families; more aborters than keepers tend to have had some type of regular employment outside the home (Evans et al., 1976). Bgljgjgn. Adolescents who are less religious are more likely to choose abortion than those who are more devout (Butts & Sporakowski, 1974; Olson, 1980). Some studies described fewer Catholics among adolescents who chose abortion (Olson, 1980). However, Eisen et al. (1983) found that abortion rates were higher among white Catholics than among white non-Catholics or Hispanic Catholics, and they concluded that the significance of religious preference as a 25 determinant of abortion is not clear. Moral or religious objections to abortion are often raised among adolescents who choose to keep their babies (Olson, 1980). Egggngngy_1n§§ntign. Adolescents who report that they wanted the pregnancy are less likely than those who say the pregnancy is unwanted to seek abortion (Fischman, 1975; Zelnik et al., 1981). This variable is described as being one of the most important factors affecting the abortion outcome (National Research Council, 1987). E .1 I . I] Several family characteristics have been found to be correlates of abortion. There is some evidence that compared to keepers, aborters are more likely to come from intact, two-parent families and to have a smaller number of siblings (Olson, 1980). Adolescents whose parents have higher levels of education are more likely to terminate their pregnancies rather than carry to term (Fischman, 1975; Zelnik et al., 1981); the latter is consistent with the finding of higher socioeconomic status among aborters. Role models within the family seem to be related to adolescent pregnancy resolution, regardless of whether the decision is to keep the baby or have an abortion. Adolescents whose family members have experience with single parenthood will be more likely to keep their babies rather than have an abortion, while adolescents whose family members have experience with abortion will more likely abort 26 (Eisen et al., 1983; Olson, 1980). Parental attitude and support of the decision are also correlated to abortion outcome. If parents’, especially mother’s, attitudes toward abortion are favorable, the adolescent will be more likely to abort than keep (Eisen et al., 1983). There is some evidence to suggest that adolescent aborters have better relationships with their parents than do adolescents who carry to term, at least in black, urban populations (Fischman, 1975). Parents reportedly have a strong influence on the decision to abort among young adolescents (Rosen, 1980). :!| E 1 . I' Adolescents who keep their babies tend to have closer and longer term relationships with the fathers of their babies, compared to aborters (Fischman, 1975). Just as with family role models, exposure to peer role models is correlated with decision outcome (Butts & Sporakowski, 1974; Eisen et al., 1983). Adolescents with single-parent friends will be more likely to become single parents rather than have an abortion, but if their friends have had abortions, they will be more likely to choose abortion than to keep. Releasers vs. Keepers The literature that contrasts adolescents who release children for adoption with those who decide to parent is sparse. This dearth of research reflects the general trend away from adoption. The studies that do exist are either 27 dated (carried out when adoption was more frequent and social norms were different) or plagued by small sample sizes (Resnick, 1984). In spite of limited generalizability due to cohort effects or sampling bias, there seem to be a few characteristics that repeatedly appear to distinguish releasers from keepers. Because the studies are not as vast as those concerning abortion, each will be reviewed individually. WWW eyer, Jones, and Borgatta (1956) did some of the earliest work on background characteristics of adolescent releasers. Their sample consisted of 52 white adolescents, 20 of whom kept their children. (Of their 48 black subjects, only 8 surrendered their babies and were not ed in further analysis.) The adolescents were recruited via their participation in a Youth Consultation Service in New York from January to June in 1954. Meyer and his colleagues identified seven variables that were associated with the decision to release: (a) religion (non- Catholic); (b) education (some college); (c) marital status of pregnancy partner (single); (d) age (under 18); (e) employment status (in school); (f) financial status (family supported); and (g) socioeconomic status (white collar, proprietary, or professional class). Among whites, predictive accuracy using a combination of the first four items yielded 83% accuracy. (Race alone resulted in 83% 28 accuracy among blacks.) Two replication studies by the same core group of authors were conducted. The first was done in 1956 on a sample of 38 white adolescents, 22 of whom opted to keep their children (Meyer, Borgatta, & Fanshel, 1959). Religion, education, age, and socioeconomic status were examined with respect to outcome decision. Only religion (non—Catholic) was clearly related to decision. In 1959, Jones, Meyer, and Borgatta (1962) investigated the relationship of religion, education, and age to outcome decision in a sample of 90 white adolescents, 71 of whom released their babies. In combination, these three variables showed some relationship to outcome, in the same directions as their earlier study. However, for this study, the cutting point for age was 17 or older; the authors did not report significance of individual variables. Vincent (1960) described background, psychological, and familial factors associated with unwed mothers’ decisions regarding pregnancy resolution. Data for this study were collected in early 1954 from a sample of women at two California maternity homes. The sample consisted of 105 white, unmarried women who were not illegitimate themselves, nor whose pregnancies were the result of rape or incest. In addition, the women in the study had not had any previous "illicit pregnancies" (Vincent, 1960, p. 112). It is noteworthy that this sample was not limited to adolescents; 29 the age range was 14 to 27 years. Respondents were given the California Psychological Inventory (Gough, 1957, cited in Vincent, 1960) and a 12-page questionnaire about their families. Vincent (1960) reported that adolescents who released their babies were slightly younger; the mean age for releasers was 19.1 years compared to 19.4 for keepers. Releasers had more education in that they completed more years of schooling, a higher percentage of them attended college, and a lower percentage of them did not finish high school, compared to women who kept. Women who gave up their babies were more likely to be in school rather than to be working compared to those who chose to parent. - w Families of women who chose adoption were more often intact, two-parent homes with fewer siblings. In addition, Vincent (1960) described the adolescent keeper as coming from a mother—dominated, unhappy home and having less favorable intrafamilial relationships. Furthermore, the keepers were reported as having more negative personality profiles. Vincent’s findings (1960, 1961) painted a picture of the“ —t as being psychologically maladjusted and less able to parent well. She was ” fl, given the norms of that time. 7. ii Q..- 30 Yelloly (1965) studied an English sample of 160 unwed mothers in 1962. She found that favorable parental attitude to the decision, no older children, and an unmarried pregnancy partner were related to the decision to keep the child. In contrast to earlier studies, there was no relationship between outcome decision and age, education, social class, family structure, or familial relationships; this discrepancy could be the result of cultural differences among samples. Festinger (1971) found her sample from their contacts with a social work agency in New York City from 1967 to 1968. She analyzed case records from 49 white and 21 black keepers and 49 white and 18 black releasers. Women in this study were all unmarried and had contacted the agency prior to the birth of their babies. This sample was not limited to adolescents; the mean ages of subjects was 21 and 19, for whites and blacks respectively. Festinger (1971) investigated several client characteristics, putative father characteristics, and aspects of the client-putative father relationship. Black women were much more likely to keep their children than white women, and most of the variables identified by this study only distinguished between outcome for whites. Compared to their counterparts who kept, white women who released their babies were found to be younger (under 20), to be students rather than working (but educational level 31 was not related), to be living with their parents during the pregnancy, to not have had prior pregnancies, to be from two—parent homes, to have parents who supported adoption, to be the same race as the putative father, and to be less involved with him. Broken parental home was more predictive of the decision when it was due to separation or divorce rather than death. For blacks, adoption was associated with younger age, parental support for the decision, and knowing the putative father for at least one year (Festinger, 1971). Grow (1969) collected data in 1966 as part of a large study based on a national sample of women in maternity homes. The sample consisted of 2,606 unwed mothers, 70% of whom were planning to release their children and 17% of whom were planning to keep (plans for the remainder were unknown). Several variables were found to differentiate the keepers from the releasers. Consistent with previous research, the releasers were more likely to be white. The white mothers who surrendered their babies were more likely to have finished high school and less likely to have come from a broken home. The putative fathers of white releasers were more available for marriage. Releasers also reported higher parental income and among those who worked, releasers held higher status jobs compared to those who kept. Most of the sample was Protestant, but in contrast to earlier studies, Grow (1969) did not find religion or age to be related to decision outcome. 32 In a more recent study, Grow (1979) interviewed 210 unmarried mothers. Data were collected in 1973 in Wisconsin. One hundred eighty-two of the women kept their babies. The sample was white, between 14 and 24 years old, and for all the women this was a first birth. Results of this study were somewhat consistent with previous findings, but there were some noted differences. Keepers and releasers could be distinguished on the basis of age, but this finding was not consistent with earlier studies: a larger percentage of women over 21 elected to release. This finding differs from earlier studies that reported age differences between groups, in which keepers tended to be older. With respect to education, similar proportions of those who never entered college kept or released their children, but releasers constituted a higher proportion of those who attended or finished college. More of the keepers came from homes in which parents separated or divorced. Other variables which distinguished between keepers and releasers were living arrangements during pregnancy (more releasers lived with parents or relatives), student vs. working status (more releasers were students), timing of request for help (keepers sought help earlier), and length of relationship with putative father (keepers knew him longer). Consistent with Festinger (1971), Grow (1979) also found that when the putative father was not the same race as the adolescent, she was more likely to keep the baby than 33 release it. There was no difference between groups, however, with respect to parental socioeconomic status or emotional adjustment (Grow, 1979). Grow (1979) suggested that releasers may represent the traditional values of previous decades, since they tend to come from two-parent homes, to grow up in smaller cities, to be less likely to consider abortion, and to attend church regularly. A study by Leynes (1980) is unique to the adoption decision literature in that it offers multivariate analysis of factors associated with adoption. There are, however, some problems with this study. Leynes (1980) used a. retrospective design using records of clients who were discharged from a Salvation Army facility that assisted pregnant adolescents to prepare for parenthood. From 1976 to 1978, 32 adolescents’ records were examined. The adolescents ranged in age from 14 to 20, with a mean of 16.69 years. They were from various racial (white, Japanese, part-Hawaiian, Mexican, Phillipino,.and mixed) and socioeconomic backgrounds. Approximately one-third of the sample had been involved with the courts for runaway or other charges, and another third were given psychiatric referrals. Twelve of the 32 women chose to release their babies for adoption. Leynes (1980) analyzed decision outcome with respect to age, socioeconomic status, parental influence, male partner influence, and level of psychiatric 34 functioning. Only two of these variables were related to outcome. Adolescents who functioned better psychologically tended to release their babies for adoption, and male partners tended to influence adolescents to keep their babies. Level of functioning was significantly correlated with parental involvement in the pregnancy, although parental involvement was not significantly related to outcome. Leynes (1980) used multiple linear regression to predict the decision, but it is unclear how this technique was appropriately used with the noncontinuous variable of outcome. ll . E l° Characteristics of unmarried women deciding to keep or release their babies for adoption have changed over the years (Grow, 1979)- w w The studies reviewed show that keepers were allegedly psychologically maladjusted, and were clearly in the minority. More than two of the major studies at that time also described“ — These findings were consistent with the then predominant conceptual understanding of adolescent pregnancy. — W By the late 1960s, the two major studies by Festinger (1971) and Grow (1969) still reported that the unmarried woman who kept her child was likely to have come from a broken home. Religion was no longer related to outcome. Findings also suggested that the relationship of age and education to outcome was no longer as clear, but when a relationship was found it was in the same direction as earlier descriptions. Neither of these studies assessed personality variables of the women keeping or releasing their babies in the late 19605. Instead of stressing the psychological basis for the decision to keep a baby, Festinger (1971) suggested a multiple-cause approach. She agreed that the decision to keep could indeed be related to the adolescent’s psychological need stemming from unstable family relationships, but it could also be related to her desire to live independently, to be a parent, and to continue a positive relationship with the baby’s father. Furthermore, awareness of poorer placement possibilities for interracial or black children could affect an unwed mother’s choice to keep. Finally, Festinger (1971) concluded that since the characteristics of keepers and releasers in the late 19605 36 did not differ significantly from earlier studies, the characteristics could not account for the increasing numbers of women who were choosing to keep their babies. Changing societal attitudes toward sexual freedom, status of women, and divorce and single-parenthood offered alternative explanations to adolescent pregnancy and the choice to keep. The most recent studies by Grow (1979) and Leynes (1980) demonstrate that the unmarried adolescent who releases her baby for adoption is now in the minority. Based on her results, Leynes (1980) concluded that patterns ‘of psychological functioning and socioeconomic status among keepers have not changed over time. However, given the sampling bias and incorrect statistical method used in her study, the validity and generalizability of her conclusions are questionable. On the other hand, the conclusions drawn by Grow (1979) are more in alignment with those of Festinger (1971). Grow (1979, p. 370) stated that: Neither social deviancy nor the psychological explanations of previous eras adequately explain why some of the pregnant unwed women of this decade decide to keep and others decide to surrender. In today’s society a social milieu has evolved that has adapted to changes in point of view regarding marriage and the family. A pregnant unwed woman coming from such a milieu will most likely keep her child. For her it is not psychologically or socially deviant. A pregnant unwed woman exposed to the traditional social milieu that adheres to older conventions is more likely to surrender her child. Similarly, she is a product of ‘her environment, and cannot be regarded as either psychologically or socially deviant. 37 In today’s society, the decision to keep the baby may represent an alternative life path for some adolescents. The decision may be rooted in social-structural forces rather than in individual personalities or moral choices. In any event, studies focusing on descriptive characteristics of women who keep or release their babies do not address the question of how the pregnancy resolution decision is made. Pregnancy resolution decision making and family influences will be reviewed in the next section. Adolescent Pregnancy Resolution Decision Making E . . || . I] Little is known as to why or how adolescents make the choices that they do about unintended pregnancies (Eisen & Zellman, 1984; Eisen et al., 1983; Evans et al., 1976; Rosen, 1980; Zelnik et al., 1981). Few studies directly address the decision making process, especially among an adolescent population. Among those that attempt to analyze factors that enter into the decision, few are theoretically driven. Haas (1974) conceptualized a general fertility decision—making model that involves three stages. It suggested that attitudes toward pregnancy and fertility goals may differ in the preconception, pregnancy, and postnatal stages. According to this model, decision making in each stage is based on the couple’s perceived susceptibility to pregnancy, perceived advantages and 38 disadvantages to childbearing, the knowledge, availability of, and acceptability of different birth-planning strategies (e.g. conception, abortion, adoption), and couple communication. While the pregnancy stage would be most relevant to pregnancy resolution decisions for adolescents, this complex model is largely theoretical, untested, and its application to the unmarried adolescent is unclear. Maskay and Juhasz (1983) proposed a ”Decision—Making Process Model” that was designed to be used for adolescent sexual decision making, including pregnancy resolution. The steps reflect a problem-solving approach: identifying the decision required and why it is needed, identifying alternative choices, anticipating negative and positive consequences of each alternative, evaluating the consequences in terms of desirability and practicality, and making a choice. This approach has value in that it may be applicable to many types of decisions, but it relies heavily on cognitive processes, and there is no evidence to suggest that this is the process by which adolescents actually reach pregnancy resolution decisions. Brazzell and Acock (1988) explored factors related to pregnancy resolution intentions. They present a conceptual model that links attitudes, adult aspirations, and significant others to adolescents’ intentions of resolving premarital pregnancy. It should be noted that although the adolescents in their sample were unmarried and sexually 39 active, they were not pregnant and faced with a decision. In this study, the model was applied to the decision to keep or abort. Their findings suggest that adolescents with positive attitudes toward abortion have stronger intentions to terminate future unwanted pregnancies. Abortion attitudes were more positive among career-oriented adolescents. The adolescents’ intentions to abort were also influenced by the perceived abortion attitudes of mothers and best friends, but mediated by the adolescent’s own attitudes. Adolescents who felt closer to their boyfriends showed less intention to terminate a future pregnancy. Lewis (1980) compared pregnancy decision making in two groups: 16 young women aged 13 to 17 (minors) and 26 women aged 16 to 26 (adults). Her subjects were unmarried and interviewed while waiting for pregnancy test results in urban pregnancy clinics. Lewis (1980) found two features that distinguished the groups. The minors expected less contact with a professional to discuss pregnancy options. The younger group also perceived their expected resolution decisions to be more externally determined (e.g. reactions of parents). The decisions in this study concerned abortion. It should also be noted that as it turned out, not all of the women in the study were actually pregnant. Most of the studies which do address pregnancy resolution decision making in adolescents have dealt with the decision to abort or carry to term. Bracken, Klerman, 40 and Bracken (1978) contended that the abortion decision is related to circumstances surrounding each individual pregnancy rather than to characteristics of the pregnant adolescents. In a series of articles, Bracken and different colleagues (Bracken, 1973; Bracken, Hachmamovitch, & Grossman, 1974; Bracken & Kasl, 1975a, 1975b) proposed a cognitive dissonance model of conflict resolution in pregnancy decisions. Four cognitive-affective stages were outlined: (a) acknowledgement of pregnancy and a corresponding positive or negative affective response, (b) generation of outcome possibilities and acceptance or rejection of each, (c) consideration of advantages of each alternative and easy or difficult decision making process, and (d) commitment to the alternative selected and a corresponding affective response to the choice. Each decision in this stage process depends on earlier stages and is meant to minimize cognitive dissonance and to maximize net gains. Bracken et al. (1978) pointed out that in their most recent study of 249 aborters and 249 matched keepers, adoption did not enter the decision—making process as a significant alternative; only 2% of the keepers were even considering this option. This is not too surprising, given that 77% of the sample was black. In the opinion of these researchers, failure to consider adoption is a reflection of subcultural values. In higher socioeconomic groups, 41 abortion is thought to replace the adoption alternative, while in lower socioeconomic groups adoption is thought to conflict with traditional values (Bracken et al., 1978). Eisen et al. (1983) addressed the decision of adolescents to abort or.carry to term, but they are unique in that their analyses were guided by a conceptual model of decision making. The model, which applies to the adoption decision as well, is: "...based on the assumption that adolescents’ pregnancy resolution decisions incorporate some form of expected utility analysis...In making a decision, the individual (and perhaps significant others) identifies the available options, weighs the perceived benefits and costs in economic, social, and psychological terms, and then chooses the one option that appears to be preferable to any of the others. While this decision making process may not be rational or logical in the cognitive developmental sense of that term..., the model does assume that the decision reached is perceived by the teenager to further certain important goals (e.g., independent living, school leaving, pursuit of further education). In this sense, the decision making is a ’rational’ process to the adolescent." (Eisen et al., 1983, p. 75) The model describes three general factors underlying pregnancy outcome decisions: psychosocial (e.g., attitudes, social relations, views of others), background (e.g., age, race, religion, educational achievement), and economic factors (e.g., time value, financial aid support). Background factors are thought to affect both psychosocial and economic factors, with economic factors also impacting background factors. Each of these three types of underlying factors also directly influences the decision to abort, 42 adopt, or keep the baby. Discriminant function analysis of the abortion decision showed that all three general factors made significant contributions to successful classification of adolescents into outcome groups (87.5% accuracy). WW Some studies have looked into the reasons why adolescents failed to choose adoption, but many of these are not methodologically rigorous. Based on her experience in a maternity home for pregnant adolescents in which an increasing proportion of clients were keeping their children, Friedman (1975) believed that more adolescents were keeping rather than surrendering in order to fulfill psychological needs. She described reasons for keeping the child as the adolescent’s way of proving her worth, fulfilling herself, gaining independence, or relieVing loneliness, boredom, or depression. Friedman’s work appears to be speculative at best, since she did not provide any data to support these conclusions. Wilt and Michael (1982) have identified several possible considerations in adolescents’ adoption decisions. Adolescents who consider adoption may feel too young for motherhood; they may not have the financial means, and/or they may feel that adoption is best for the child under the circumstances. However, they may also believe that relinquishing the baby is selfish and cruel; that their parents or boyfriend may attempt to gain custody of the 43 child; that adoptive parents may not treat the child well; that procedures used by adoption agencies, courts, and judges are confusing and fear provoking; and that feelings of guilt and regret may develop in the future. Addressing some of the same considerations, Richards and Willis (1983) describe the adoption decision making process in terms of four major decisions: (a) whether to let the baby be adopted at all, (b) who should adopt the baby, (c) at what point to agree to the adoption, and (d) whether the adoption should be an agency, family, or private adoption. Musick et al. (1984) did a study to determine why adoption is so infrequently chosen by adolescents. They interviewed pregnant teens and young mothers at community services agencies targeting prevention of adolescent pregnancy. Respondents were white, black, and Hispanic, but the authors do not report ages or sample size. According to this study, the adolescents identified adoption as an alternative, just not one that they would choose. Among this group, however, abortion was given the same response. These adolescents perceived not keeping the baby as personal failure. In addition, there were few familial or peer sanctions against keeping and little support for adoption. Families and boyfriends were found to have raised opposition to adoption when it was considered. Some adolescents expressed fears that if the child were released, it would not be placed in a proper home. Adolescents who were known 44 by the respondents to have released babies were described as having specific life goals. Musick et al. (1984) also noted that adolescents did not discuss pregnancy resolution alternatives in terms of the future life of the child. Very few focused on the implications of different alternatives on the child’s life, a finding consistent with Lewis (1980). Finally, the adolescents’ inability to project into the future and establish realistic future goals was seen as a barrier to adoption. Somewhat in contrast to Musick et al. (1984), Barth (1987) found that young women were concerned about the impact of the decision on the child and were uncertain about the baby’s future. They tended to prefer adoptions that involve choosing among potential adoptive families and viewing videotapes of couples as they described themselves. An expressed preference for reducing uncertainty is suggestive of the importance of adoption options that encourage continued contact with the child. A greater number of adoption options may increase the likelihood that young women will choose adoption, and open adoption practices may be especially attractive to young women who are concerned about the future of their children. MW Many studies have reported that mothers have an impact on the pregnancy resolution decisions of their adolescent daughters (Bracken et al., 1978; Eisen et al., 1983; Eisen & 45 Zellman, 1984; Hudis & Brazzell, 1981; Musick et al., 1984; Ortiz & Nuttall, 1987; Rosen, 1980; Rosen, Benson, & Stack, 1982; Young et al., 1975). Parents are a primary reference group for most adolescents and not surprisingly, mother’s attitudes about alternatives are related to outcome decision (Eisen et al., 1983; Hudis & Brazzell, 1981; Musick et al., 1984; Young et al., 1975). When mother was supportive of the adolescent’s decision, the adolescent was more likely to have reported that the decision was less conflicted (Bracken et al., 1978). Decision satisfaction was also positively related to strong support from mother, especially among keepers (Eisen & Zellman; Ortiz & Nuttall, 1987). Not many studies address mothers’ reactions to their adolescent daughters’ pregnancies. Mothers seem to initially respond negatively to the pregnancy (e.g., shock, anger, disappointment, worry), but learn to accept it (Musick et al., 1984; Smith, 1975) and help their daughters plan for the baby (Young et al., 1975). Parental involvement with the pregnancy may have been assumed prior to the legalization of abortion, when the adolescent was faced with few, if any, alternatives (illegal abortion or running away) in order to keep her pregnancy secret. Since the legalization of abortion, however, adolescents have had more choice in whether or not to disclose a pregnancy to their parents (Rosen, 1980). Rosen (1980) addressed the question of whether or not adolescents 46 involved their parents in pregnancy resolution decision in a study she conducted of 432 adolescents. Adolescents were 18 or younger and unmarried when they conceived. Two hundred and fifty of them were white, and 182 were black. Data were collected in 1974 to 1975 using a stratified sampling of organizations in Michigan that provided services to women with unwanted conceptions. Rosen (1980) inquired about the involvement and influence of parents, friends, and pregnancy partner on the decision. She analyzed the data according to the following groups: white keepers (101), white aborters (105), white releasers (43), black‘keepers (136), and black aborters (46). Rosen (1980) found that most adolescents did not involve either parent until they knew for sure that they were pregnant. Once the pregnancy was confirmed, however, over half of the adolescents in each category went to their mothers. Mother’s influence was considerable for all groups, but it was greatest among the releasers and least among the white keepers. The white keepers were influenced primarily by their pregnancy partners. For all other groups, mother’s influence was greater than father’s, friend’s, or pregnancy partner’s. In addition, for the white adolescents, mother’s influence was inversely related to the adolescent’s perceived competence, and for all groups it was positively related to conflict. Rosen (1980) has interpreted these results to mean that adolescents who 47 experienced conflict and perceived themselves to be less competent were more likely to turn to their mothers. The type of influence that parents exerted was investigated by another study (Rosen et al., 1982). The sample for this study was rural, primarily white women under the age of 20, who sought pregnancy services from private physicians. Two-thirds of the sample of 100 chose to have an abortion, while 12 kept the baby and stayed single, 17 kept the baby and got married, and 1 kept the baby but was already married. Rosen et al. (1982) found that direct pressure, defined as the potential or actual use of resources to gain compliance, was rarely felt by adolescents, regardless of the decision. When it was perceived, the pressure was in the direction of abortion. Adolescents who chose abortion more commonly reported indirect pressure from parents. That is, they expected that resources would be used and took action to avoid that outcome. Direct influence, parental advice and support regardless of chosen outcome, was experienced by about one-quarter of the aborters and nearly one-half of the keepers. It was most common in keepers who stayed single, but this finding is tentative due to the few adolescents who chose this outcome. Indirect influence was defined as situational factors not directly related to decision-making, such as teenage pregnancy among mothers. This category seemed to have greatest impact on adolescents who chose to 48 marry and keep their children. Furstenberg (1981) holds the opinion that more research needs to be conducted concerning the role of the family in adolescent pregnancy. He stated that almost no research is available concerning the link between early child bearing among adolescents and their family organization (Furstenberg, 1981). Furthermore, he believes more information is needed about when the adolescent tells her family and how their responses affect her decisions. Although the work of Rosen and her colleagues (Rosen, 1980; Rosen et al., 1982) has shed much light on mother’s influence, not much is known about mother’s involvement in the adolescent’s pregnancy or in the decision to release the baby. Purpose of This Research This study will compare a group of adolescents who released their babies for adoption with a group who chose to keep their children. The purpose of the study is three- fold. First, it will provide a current description of the characteristics of the adolescent releasers. Comparisons to earlier trends can be made. The current description will provide new information about the adolescents as well, namely attitudes toward abortion and adoption, and more details about family characteristics. Second, this study will explore the course of the adolescents’ mothers’ involvement in the pregnancy. Not much is currently known 49 about how keepers and releasers might differ with respect to their mothers’ involvement in the pregnancy or their influence on the decision. Third, the relationship of family characteristics and mother’s influence, as they relate to decision outcome, will be addressed. The larger picture of family context and how it may affect both mother’s influence and the pregnancy resolution outcome has not been well researched. Knowledge of the adolescent who releases her baby and how family factors and mother’s influence are involved in the decision will be useful not only to clinicians who provide pregnancy counseling services to individuals, but also to policy makers who address the societal problem of adolescent pregnancy. The effect of individual, family, and pregnancy—related variables on the decision will be conceptualized according to the decision making model proposed by Eisen et al. (1983). Application of the model in this study is illustrated in Figure 1. This study will focus on background and psychosocial factors, with an emphasis on the family. Eisen et al. (1983) did not include in their discussion how future aspirations might be included in this model. However, future goals, such as educational and occupational plans, may be thought of as economic factors in that they do depict the adolescent’s values as to how to spend the limited resources of time and energy. The present study will not test this model, but the model does serve as 50 a framework for conceptualization. UNDERLYING FACTORS W1 Sachem E;__-*oncm‘~ c abortion attitudes age time adoption attitudes SES value: pregnancy living educational future arrangements f_" achievement 9’ goals self discovery of employment experience the pregnancy pregnancy history mother’s discovery family structure of the pregnancy family environment mother’s influence perception of own influence perception of mother’s influence L 4.1 Abort DECISION Adopt Keep Figure 1 Application of Pregnancy Decision Making Model (Eisen et al., 1983) to This Study 51 Hypotheses The following specific hypotheses will be tested: fiypgtn§§1§_gl: Adolescents who release their babies for adoption will be younger than those who keep their babies. More recent findings suggest a less clear relationship of age to outcome than early studies of releasers (Grow, 1979), but most of the studies described the releaser as younger. Many studies, however, were not strictly limited to an adolescent sample. It is anticipated that adolescent releasers will be younger than their keeper counterparts. flxpgtn§§j§_gz: Abortion attitudes will not be related to outcome group. In studies about the abortion decision, keepers often raise moral or religious objections to abortion (Olson, 1980). Although no study directly reports on abortion attitudes among releasers (probably because abortion was not a legal alternative at the time these studies were done), Grow (1969) suggested that releasers have more traditional values. Given this, it is expected that attitudes toward abortion in each group will be conservative and would explain why neither group of young women chose to terminate their pregnancies. No differences are expected between groups because they all rejected abortion as an alternative. 52 H1291h§§i§_£§= Releasers will have higher future educational and occupational goals as compared to keepers. Previous studies comparing releasers and keepers found consistently that the releasers tended to be in school rather than working (Festinger, 1971; Grow, 1979) and more likely to have completed more education compared to keepers (Grow, 1969, 1979). It is expected that similar to the literature on aborters, releasers will have higher future aspirations compared to keepers. uxngtn§§j§_gg: Releasers will be more likely than keepers to have grown up in a two-parent family. Based on past findings of the adoption literature, releasers are exoected to have grown up in nonbroken homes, as compared to keepers. That broken homes are associated with the adoption decision outcome is one of the most consistent findings in the literature. flxngtn§§j§_£§: Family environment will be related to outcome group. There is no previous research on which to base a prediction concerning how family environment will affect outcome. However, the adolescent makes her decision in the larger context of the family, and intuitively it makes sense that the type of context will be related to outcome. flxpgtn§§i§_£§: Family environment will be related to the adolescent’s-abortion and adoption attitudes, regardless of decision outcome. It is expected that one way in which family environment will impact the decision is through its effects on 53 attitudes. This is one example of how background factors may influence psychosocial factors in the decision making model (Eisen et al., 1983). There is no specific literature on family environment and attitudes, but parental communication of values in the context of the family is one way in which children are socialized (Moore et al., 1986). uxggtng§i§_£1: Family environment will be related to the adolescent’s occupational and educational plans for her future, regardless of outcome. Family environment is also expected to impact the final decision by its influence on the adolescent’s future aspirations. Adolescents raised in families which promote intellectual achievement, for example, would be expected to have higher future aspirations than if these values were not held by the family. flyggtne§1§_£§: Family environment will be related to the adolescent’s perception of influences on her decision. Family environment is also expected to be related to how the adolescent views herself and others with respect to the influences on her decision. In families which stress autonomy and independence, adolescents could be anticipated to perceive their decisions as more autonomous than the adolescents raised in a different type of family environment. 54 flxngtne§i§_£2: Mother’s influence will be related to outcome. Based on the work of Rosen and her colleagues (Rosen, 1980; Rosen et al., 1982), it is anticipated that the help that mothers offer the adolescents will be related to the final outcome. Although no research is available, mothers of releasers are expected to offer less help to their daughters. uypgtng§i§_glgz Perception of mother’s influence will be related to mother’s discovery of the pregnancy and living arrangements during the pregnancy. Adolescents are expected to report that the earlier their mothers found out about the pregnancy, the greater degree of influence she had on the decision . Similarly, if the adolescent lived in the parental home during her pregnancy, it is expected that she will report that her mother influenced her decision to a greater degree. These relationships are expected to be the same for both keepers and releasers. CHAPTER THREE METHODOLOGY OVERALL STUDY Design This study compared a group of adolescents who released their babies for adoption with a group of adolescents who kept their children along several individual and family variables, and examined the influence of family variables on the decision. The data were collected as part of a larger study, "Adoption Decisions: Personal and Social Context", funded by the Office of Adolescent Pregnancy Programs. The study was conducted through the Department of Pediatrics and Human Development at Michigan State University from 1984 to 1987. The study design is retrospective. Because of constraints imposed by the funding agency, data collection could not begin until after the birth of the baby, and in the case of the young women who released their babies for adoption, until after the formal court date confirming the release. new figgngcs. The final criteria for inclusion in the study for adolescents who kept their babies were (a) the adolescent was l9 or younger at the time of conception; (b) the adolescent was unmarried at the time of conception; and (c) the adolescent had no older children at home. 55 56 Adolescents were recruited in Flint and Grand Rapids, where the cooperation of local hospitals was obtained. Adolescents were approached by staff during their hospital stays following delivery, and invited to participate in the study. Each adolescent was offered $10.00 to participate, and approximately 50% of the adolescents who were approached agreed to take part. Written parental consent was obtained for adolescents who were minors. Examples of the contact letters and consent forms that were used are in Appendix A. Adolescents who agreed to participate were recontacted once they returned home, and an interview appointment was scheduled. Individual interviews were conducted three to six weeks after delivery by trained research staff. The interview consisted of both open- and close-ended questions about the adolescents’ pregnancy experiences, attitudes and beliefs, and demographic information. The interview lasted approximately one and one-half hours and was conducted in a place selected by the adolescent that would ensure convenience and privacy. Most of the interviews were done in the adolescents’ homes, with only a few conducted in other locations such as the hospital cafeteria or a local McDonalds. A few additional adolescents were recruited through the local school for pregnant adolescents in Flint. With the exception of the recruitment site, the same recruitment and interview procedures were followed with these adolescents. 57 A total of 180 adolescents who kept their children were interviewed. Belggfiggs. The recruitment procedures described above yielded only two adolescents who released their babies for adoption. Releasers were primarily recruited by adoption agencies throughout the state. Recruitment was difficult due to lack of agency cooperation (refusal was usually on the grounds that the interview would be upsetting for their adolescent clients) and lack of adolescents who met the inclusion criteria for the study. In hopes of increasing sample size, the criteria for inclusion in the study were slightly modified for releasers. The final criteria were: (a) the adolescent was 19 or younger at the time of conception; (b) the adolescent was unmarried at the time of conception; (c) the adolescent released her baby within three months of birth; and (d) there were no restrictions on the adolescent having older children at home. In addition, the agencies searched their records for adolescents who met these criteria and relinquished their children in the prior 18 months. Examples of the contact letters and consent forms used by the adoption agencies are in Appendix A. Newspaper advertisements were also placed, but produced no results. A total of 17 adolescent releasers were interviewed. The majority of these (n=11) were recruited through Catholic Social Services in Ann Arbor, an agency which offers various degrees of openness in adoption 58 placement procedures. The interview for releasers was essentially the same as that used for keepers, with appropriate modifications in wording to accommodate for decision outcome. PRESENT STUDY Sample Balsam This study included 14 white releasers. Of the total 17 adolescent releasers interviewed, 14 were white, 1 was black, and 2 were of other races. Race is a significant variable in adoption that needs to be controlled; due to insufficient sample size to include race as an independent variable, nonwhite respondents were excluded. All of the 14 white releasers were unmarried at the time of the interview and had no older children at home. Race, number and ages of children, and marital status have been identified as among the factors most crucial for matching groups of releasers and keepers in adoption research (Resnick, 1984). Keenena Seventy-nine of the 180 adolescent keepers interviewed were white, but sample size of keepers for the present stucy is limited by the small number of releasers. In an attempt to increase the sample size yet retain relatively equal group sizes for comparison, the white keepers were further broken down into four subgroups based on the outcome alternatives they considered when making their decisions. 59 This also allowed for comparisons between adolescents who considered similar options but chose different outcomes. These groups are: "Pure Keepers", who considered keeping as their only option (n=44); "Aborting Keepers", who considered abortion as the only alternative in addition to keeping (n=18); ”Adopting Keepers", who considered adoption as the only alternative in addition to keeping (n=12); and "Deciding Keepers“, who considered all three alternatives (n=5). The last group were not included due to insufficient number for meaningful statistical comparisons. A random sample of 14 adolescents were drawn from each of the groups (with the exception of the Adopting Keepers, whose size is limited to 12). Thus, the sample of keepers for this study was "Pure Keepers" (n=14), "Aborting Keepers" (n=14), and” Adopting Keepers" (n=12). The keepers in these groups were matched with releasers with respect to marital status at the time of the interview and number of older children. The overall sample used for this study is described in Table 2. Variables and Measurement The variables examined in this study were categorized along three dimensions: individual variables, family variables, and pregnancy-related variables. A description of the specific variables that were included is given below. Data for this study were obtained by the adolescent’s self- report during the interview. The exact interview questions and response categories for each variable are given in 60 Table 2 Description of Sample * Ema Outcome N W Pure Keepers Kept 14 Only keeping considered Aborting Keepers Kept 14 Only abortion & keeping considered Adopting Keepers Kept 12 Only adoption & keeping considered Releasers Released 14 (insufficient sample to control for alternatives considered) * Adolescents in each group are white, unmarried, and do not have older children Appendix B; these questions are a subset of the larger interview schedule. I I' 'I 1:, . I] W The adolescent’s age at the time of the interview, the highest grade she had completed at the time of the interview, and her religious preference were each obtained by single questions during the interview. Employment. Of interest was whether or not the adolescent ever held a job. This was also determined by a single interview question. Enggngngx_n1§§g£1. Two aspects of pregnancy history were measured. These were whether or not this was the adolescent’s first pregnancy and if not, what were the 61 outcomes of any previous pregnancies. Abggtign_attitgge§. Mirande and Hammer (1974) developed a 15-item Guttman scale they called Premarital Abortion Permissiveness (PAP). The scale was designed to assess under what conditions single women would view abortion as an alternative to unwanted pregnancy. Responses are scored on a 6-point Likert scale from strong agreement to strong disagreement. (See Appendix 8). Five items from this 15-item Guttman scale were selected for the interview schedule. Time constraints prevented inclusion of all 15 items. The items were reworded from first person to third person. Scores on the shortened abortion scale range from 5 to 30 points. A high score indicates the adolescent is opposed to abortion, while a low score indicates there are some situations in which she would consider it was acceptable for a woman to have an abortion. Adontign_gttitug§§. Adolescents’ attitudes toward adoption were characterized along two dimensions: ”Differ" and "Afraid". These scales were developed for the larger study. Factor analysis of the items based on a college student pilot study determined groupings of the items, which are found in Appendix B. The concept for each scale was derived from Kirk’s (1964) work with adoptive parents, which focused on how the adoption was handled within the immediate family, by friends, and by extended family members. The 62 experience of adoptive parents revealed two general perceptions about how adopted children were considered to be different from nonadopted children: (a) The adopted child was expected to behave differently from a ”natural" or nonadoptive child (i.e., to be deviant or more problematic), and (b) more difficulties were expected in the integration of the adoptive child into the family, due to either the child’s behaviors or the family’s reactions to the child. "Differ" is a scale that refers to what degree the adolescent perceives adopted children to be different from nonadopted children. It is based on five items, and responses are scored on a 4-point Likert scale from "strongly agree" to "strongly disagree". Scores range from 5 to 20 points, with a high score indicating that the adolescent believes adopted children are not different from nonadopted children._ As part of the analysis of the larger study, Cronbach’s alpha statistic of reliability was computed for "Differ” on a subsample of 105 adolescent keepers as .67 (Kallen, Popovich, Griffore, & Powell, 1988). "Afraid” is a scale that ascertains whether the adolescent believes an adopted child will have trouble fitting into an adoptive family. This scale was measured by six items. Responses were scored according to a 4-point Likert scale from "strongly agree" to "strongly disagree”. Scale scores for "Afraid" range from 6 to 24 points. A high score suggests the adolescent is not worried that an adopted 63 child will fit into an adoptive family, while a low score indicates that she does have concerns. Cronbach’s alpha for "Afraid" based on the same subsample of 105 adolescent keepers was .74 (Kallen et al., 1988). Eytu;g_ggg1§. The adolescent was asked about her plans for finishing or continuing her education, what kind-of job she would like to have in the future as well as what she thinks her chances will be of getting it. Single items in the interview schedule will assess these goals. E .1 H . I] Sggjggggngmig_§;§tg§. Pilot interviews for the larger study indicated that adolescents were unable to accurately report family income, and it was not asked in the interview. Socioeconomic status was estimated by mother’s education. Mother’s education was chosen because not all families are expected to be two-parent families. Although mother’s occupation at the time of the pregnancy was assessed, mother’s education rather than occupation is the preferable estimate of socioeconomic status because not all mothers were necessarily employed outside the home. Mother’s education is the proxy for socioeconomic status used in other studies of adolescent pregnancy (e.g., Zelnik et al., 1981). Eamilx_§tnugtune. Several aspects of family structure were measured. These variables were mother’s age at time of interview, previous occurrence of adolescent pregnancy in 64 the family, and composition of household while the adolescent was growing up. The latter includes number of parents (single- or two-parent home), number of siblings, and nuclear vs. extended family structure. Egmilx_§nxicgnm§n§. Family environment was measured by the Family Environment Scale (FES), Form S (Moos, Insel, & Humphrey, 1974). Form S is the short form of the FES and uses the first 40 items of the 90-item FES (Form R). It was chosen because of constraints on interview length. The actual items are given in Appendix 8. FES measures the social climate of families, focusing on family functioning along three general dimensions: Relationship, Personal Growth, and System Maintenance (Moos et al., 1974). Family relationships are measured by the Cohesion, Expressiveness, and Conflict subscales. Personal Growth is assessed by the Independence, Achievement Orientation, Intellectual-Cultural Orientation, Active Recreational Orientation, and Moral-Religious Emphasis subscales. Finally, the System Maintenance Dimension is characterized by the Organization and Control subscales. Each subscale in the short form consists of four true-false items. A brief description of each of these subscales is given in Table 3. Data on the psychometric properties of the FES that are reported are based on the 90-item version, Form R. The psychometric properties, especially reliability, are good 65 for the 90-item version of the FES and make it one of the best measures available for assessing family environment (Busch-Rossnagel, 1985; Moos et al., 1974; Moos & Moos, 1976). Very limited data are available for Form S (Mitchell, 1985). In the preliminary manual, Moos and his colleagues (1974) reported that Form S and Form R subscale correlations were .80 and above, but these data were based on a small sample of 11 families. No further psychometric data on Form 8 have since been published. Normative data, however, are available for the short form of the FES; they are based on the same normative sample as the normative data for Form R (Moos et al., 1974). This preliminary normative sample consisted of 285 families from a variety of socioeconomic levels and family sizes (Moos et al., 1974). Standardized subscale scores are based on this normative sample. Moos and Moos (1976) have further used the FES to create typologies of family environment. Cluster analysis of standard subscale scores in a heterogeneous sample of 100 families yielded six distinct typologies. These typologies are Expression-Oriented, Structure—Oriented, Independence- Oriented (with subtypes Expressive—Independence and Structured-Independence), Achievement-Oriented (with subtypes Achievement via Independence and Achievement via ConformitY). Moral/Religious Oriented (with subtypes 66 Table 3 Family Environment Subscale Descriptions Cohesion Expressiveness Conflict Independence Achievement Orientation Intellectual- Cultural Orientation Active Recreational Moral-Religious Emphasis The extent to which family members are concerned and committed to the family and the degree to which they are helpful and supportive to each other. The extent to which family members are allowed and encouraged to act openly and to express their feelings directly. The extent to which the open expression of anger and aggression and generally conflictual interactions are characteristic of the family. E 1 i II E' . The extent to which family members are encouraged to be assertive, self- sufficient, to make their own decisions, and to think things out for themselves. The extent to which different types of activities (e.g. school and work) are cast into an achievement-oriented or competitive framework. The extent to which the family is concerned about political, social, intellectual, and cultural activities. The extent to which the family participates actively in various recreational and sporting activities. The extent to which the family actively discusses and emphasizes ethical and religious values. 67 Table 3 (cont’d.) . E' . Organization The extent to which order and organization are important in the family in terms of structuring of family activities, financial planning, and the explicitness and clarity of rules and responsibilities. Control The extent to which the family is organized in a hierarchical manner, the rigidity of rules and procedures, and the extent to which family members order each other around. Nete. From ' in u l for famil v' hm f so l work enyironment ecele, egg grgug envjcenment ecele (Table 1.1, p. 4) by R.H. Moos, P.M. Insel, and B. Humphrey, 1974, Palo Alto: CA, Consulting Psychologists Press, Inc. Reproduced by special permission of the Publisher, Consulting Psychologists Press, Inc., Palo Alto, CA. Further reproduction is prohibited without the Publisher’s consent. unstructured and structured), and Conflict—Oriented (with subtypes unstructured and structured). Each type is associated with a particular pattern of subscale scores. Moos and Moos (1976) asserted that different social environments in the family may be linked to different family outcomes, and these typologies may be a useful way to conceptualize families for research. WWW Enegnene1_1ntentiene. The adolescent’s intention of getting pregnant in her relationship with the pregnancy partner was also assessed. It was measured by a single item on the interview schedule. 68 §el£_g1§eexenx_e£_enegnenex. Two variables fall into this category: how far into the pregnancy the adolescent was when she realized she was pregnant and her reaction to being pregnant. Both were measured by individual items on the interview schedule. Wang- Three variables relevant to mother’s discovery of the pregnancy were examined. These are: when she found out, how she found out, and how she felt about her daughter’s pregnancy. Liying_e;;engemen§e. Where the adolescent lived during her pregnancy was determined by a single question on the interview schedule. Of primary interest is whether or not she lived with her family of origin. §elf_inilgenee_en_the_geeieien. Self influence refers to what the adolescent wanted to do to resolve her pregnancy, specifically, the alternatives that she considered before making her final choice. Alternatives considered is a variable that was controlled in this analysis. This allowed for more subtle comparisons between adolescents who considered different options but made different decisions. Interview questions assessed whether the adolescent considered abortion, adoption, or keeping her baby. MetneLLe_jnilgenee_en_1he_fieeieien. Mother’s influence was assessed by a number of items. These included: mother’s preference for outcome, resources she offered the 69 adolescent, and resources the adolescent wanted from her that were not offered. EeLeeetien_e£_eel£_in£lgenee. The adolescent’s perception of the degree to which her final decision was her own was assessed by a single question in the interview schedule. 81W Mother’s perceived influence was measured by several questions. These involved the adolescent’s assessment of what resources her mother offered and how her help or lack of it affected the ease and outcome of her decision, how she thought mother acted toward her during the pregnancy, and to what degree she believes mother influenced her decision. Qgteeme. Outcome is measured by the adolescent’s final decision to keep or release her baby for adoption. The keeper subgroups will be treated as separate outcomes. The variables used in this study are summarized in Table 4. Table 4 Summary of Variables INDIVIDUAL VARIABLES Adolescent’s age at time of interview Education Religion Employment Pregnancy history Abortion Attitudes Adoption Attitudes Differ Afraid 70 Table 4 (cont’d.) Future goals education occupation FAMILY VARIABLES Socioeconomic Status (mother’s education) Family Structure mother’s age number of siblings family history of adolescent pregnancy nuclear vs. extended single- or two-parent household Family Environment Scale (Form S) PREGNANCY-RELATED VARIABLES Pregnancy intentions Self Discovery of Pregnancy when pregnancy was realized adolescent’s reaction to pregnancy Mother’s Discovery of Pregnancy when mother found out how mother found out mother’s emotional reaction when she found out Living arrangements during pregnancy Mother’s Influence on Decision mother’s preference for outcome resources mother offered resources adolescent wanted that were not offered Perception of Self Influence degree to which decision was adolescent’s own decision Perception of Mother’s Influence effect of resources offered on decision effect of resources not offered on decision mother’s actions during pregnancy degree of mother’s influence on decision Outcome outcome group (keeper subtypes and releasers) CHAPTER FOUR RESULTS The findings of this study are presented in two sections: Description of the Sample and Analyses of Hypotheses. Variables not used in the specific hypotheses are included in the sample description. A significance level of .05 was used throughout. SAMPLE DESCRIPTION Given the differences in recruitment procedures and sites for the adolescents who kept (n=40) and those who released their babies (n=14), description of the sample was analyzed separately for each. Analysis for variables in this section was also done by outcome group. No differences were found between "Keepers" and "Releasers" or between the four outcome groups unless specifically mentioned. W Age_eng_neee. All the adolescents in the sample were white and had never been married. The mean age of the adolescents was 17.1 years; there was no age difference between adolescents who kept their babies and those who released them for adoption (see Hypothesis #1 for analysis of age by outcome group). Egueetien. The adolescents’ educational status ranged from completion of seventh grade to having two years of college. Approximately two-thirds (66.7%) of the sample had not yet completed high school at the time of the interview; 71 72 27.8% were high school graduates, and 5.6% completed some education beyond high school. All the adolescents were grade appropriate for age. Belieien. Religious preference did vary between ”Keepers" and "Releasers" (Chi-Square=18.60, df=4, p=.0009). Among the adolescents who kept their babies, 50.0% were fundamentalist Protestant; 22.5% identified themselves as liberal Protestants, and 20.0% were Catholic. The remainder of the "Keepers" claimed no religious affiliation (5.0%) or some other religious belief (2.5%). In contrast, none of the "Releasers" belonged to fundamentalist Protestant groups; one-half identified themselves as Catholic, and just over one-third (35.7%) had no religious preferences or beliefs. These data are shown in Table 5. Outcome groups also differed significantly with respect to religious affiliation (Chi-Square224.79, df=12, p<.02), and these differences appear to be accounted for by the same patterns between keepers and releasers already described. Emelexment. Outcome groups did not differ in employment experience. Nearly three-quarters (72.2%) of the sample responded positively when asked if they had ever held a job. EaijLBaclsamund Stcgefiuge. Both "Keepers" and "Releasers" were raised in nuclear rather than extended family systems; only two adolescents (3.7%) reported having extended family members 73 Table 5 Religious Affiliation by Final Decision Protestant Protestant QLQHE N. Lingual £9.09... Wigwam Pure Keepers 14 28.6% 35.7% 28.6% 7.1% 0.0% Aborting Keepers 14 21.4% 57.1% 21.4% 0.0% 0.0% Adopting Keepers 12 16.7% 58.3% 8.3% 8.3% 8.3% Releasers 14 14.3% 0.0% 50.0% 35.7% 0.0% Chi-Square=24.79 Df=12 p<.02 living in the home while they were growing up. The relationship between single or two-parent families and outcome group is presented in Hypothesis #4. Families of adolescents who kept (all types) and who released their babies also did not differ with respect to family size. The mean numbers of sisters and brothers reported in the family while the adolescents were growing up were 1.15 and 1.19, respectively. The majority of the adolescents’ families of origins (77.8%) did not have a history of family members getting pregnant prior to high school graduation. Early pregnancy in the family was not related to outcome group or the adolescent’s final choice. 74 Mfllhfiifi- No differences existed in the reported ages of the mothers of the adolescents by final decision or outcome group. Mothers ranged in age from 31 to 53, with a mean age of 41.1 years. Mothers’ educational attainment also did not differ between groups. About one-fifth of the mothers (20.4%) had less than a high school education, compared to 44.4% who had graduated from high school and 35.2% who had some education beyond high school. The majority of mothers (63.0%) were working outside the home when their daughters became pregnant. However, as seen in Table 6, more mothers whose daughters chose to release their babies for adoption (92.9%) were employed at that time, compared to mothers of "Pure Keepers“ (42.9%), "Aborting Keepers" (57.1%), or "Adopting Keepers" (58.3%) (Chi-Square=8.11, df=3, p<.05). There were no differences between the mothers with respect to the type of jobs they held. Of the mothers who were employed when their daughters got pregnant (n=36), 22.2% were in professional or executive positions; just under one third (30.6%) worked at technical, sales, or clerical jobs; another 27.8% held service type jobs, and the remainder worked in craft or repair (8.3%) or in operatives (8.3%). Occupations were coded according to the Census of Population, Classified Index of Industries and Occupations (U.S. Department of Commerce, 1980). 75 Table 6 Mother’s Employment Status at Time of Pregnancy Mother Mother Row Group N Wegkjng Net Werking legal Pure Keepers 14 42.9% 57.1% 100% Aborting Keepers 14 57.1% 42.9% 100% Adopting Keepers 12 58.3% 41.7% 100% Releasers 14 92.9% 7.1% 100% Chi-Square=8.12 Df=3 p<.05 W Egegnenex_njetenx. For the majority of adolescents (84.9%), this was a first pregnancy. The number of adolescents who had been pregnant before (n=8) was too small for further statistical analysis. However, none of the previously pregnant adolescents kept the baby: One "Adopting Keeper” had released her older child for adoption; 5 had abortions (4 “Aborting Keepers“ and 1 ”Releaser”), and 2 had miscarriages (1 "Pure Keeper" and 1 "Adopting Keeper"). Egegnenex_in1entiene. Few of the adolescents (15.1%) were deliberately trying to conceive. Just over half (52.8%) reported they were not thinking about the possibility of pregnancy, whereas 32.1% were trying to keep from getting pregnant. One "Aborting Keeper" refused to answer the question. The adolescents’ pregnancy intentions were unrelated to outcome group or final choice. 76 MW- Adolescents did differ with respect to how far they were into the pregnancy when they first suspected that they might be pregnant. When all types of "Keepers“ were compared to the "Releaser" group, a two-tailed t-test showed this difference was statistically significant (t=-2.87, df=52, p=.006). "Releasers" were an average of 2.86 months into the pregnancy, compared to "Keepers" who suspected pregnancy at 1.63 months. These data are presented in Table 7. Analysis of variance of outcome groups, however, was just short of significance at the .05 level (F22.71, df=3, p=.055). Table 7 Timing of Pregnancy Realization by Decision Outcome Mean Standard Standard m N m Malian Em All Keepers 40 1.63 .93 .15 Releasers 14 2.86 2.25 .60 E 1 | I . E . 2-Tailed 3431.549 9i mm -2.87 52 .006 When they realized the pregnancy, “Pure Keepers” were more likely than other groups to describe themselves as feeling clearly positive about the pregnancy (e.g., "happy", 77 "thrilled", "excited"). “Releasers" and ”Adopting Keepers", on the other hand, more often tended to react negatively when they realized they were pregnant (e.g., ""terrible", "disappointed at myself", "empty and cold, scared") or to have mixed feelings (e.g., "upset but happy", ”kind of happy in a way and sad in another way"). These data are presented in Table 8. Table 8 Adolescents’ Reactions by Outcome Group Group H mm W Mixed 9523.: Pure Keepers 14 64.3% 21.4% 14.3% 0.0% Aborting Keepers 14 21.4% 42.9% 21.4% 14.3% Adopting Keepers 12 0.0% 58.3% 41.7% 0.0% Releasers 14 0.0% 64.3% 35.7% 0.0% Chi-Square=28.17 Df=9 p<.001 MQLDQLL§_Leg§tien. Most mothers (59.3%) felt negatively about their daughter’s pregnancy. Only a few (9.3%) were pleased by the news or had mixed feelings (3.7%). Just over one-quarter of the adolescents (27.8%) reported that either they did not know how their mothers felt or that she responded in some other way. Mother’s reaction to the pregnancy was unrelated to the adolescent’s final decision or to the options she considered. 78 ANALYSIS OF HYPOTHESES fixeetnee1e_£1: Adolescents who released their babies for adoption will be younger than those who kept their babies. This hypothesis was not supported among the adolescents in this sample. One-way analysis of variance (ANOVA) of age at time of interview by outcome group was not significant (F=.73, df=3, p<.54). The mean ages for each group ranged from 16.5 to 17.4 years, with an overall mean of 17.1 years. fixeetnee1e_£2: Abortion attitudes will not be related to outcome group. Abortion attitudes were measured by a shortened version of the Premarital Abortion Permissiveness (PAP) scale (Mirande & Hammer, 1974). Scores ranged from 7 to 29 (with a high score indicating opposition to abortion), showing variability in abortion attitudes within the sample. Cronbach’s alpha (Norusis, 1988) for the scale was calculated as .72. Abortion attitudes were, in fact, related to the adolescent’s decision about pregnancy resolution. Table 9 shows that one-way ANOVA of abortion attitudes by outcome group was significant (F=5.56, df=3, p<.002). Pairwise comparisons of group means using the Scheffe method (Norusis, 1988) indicated that "Aborting Keepers" (mean score of 16.21) were significantly more liberal in their abortion attitudes compared to "Adopting Keepers" (mean score of 23.25). "Pure Keepers" and ”Releasers" were 79 between these two groups, with respective mean scores of 20.21 and 20.57. Table 9 Abortion Attitudes by Outcome Group Sum of Mean Source Di W 5.993.135. E 2 Between Groups 3 332.44 110.81 5.56 .002 Within Groups 50 996.39 19.93 Total 53 1328.83 QLQHQ fl M§§n_§§Q£§ * Pure Keepers 14 20.21 Aborting Keepers 14 16.21 Adopting Keepers 12 23.25 Releasers 14 20.57 * "Aborting Keepers” and "Adopting Keepers" differed at p<.05 using Scheffe pairwise comparisons. No other groups significantly differed from each other. fixeetnee1e_£3: Releasers will have higher future educational and occupational goals as compared to keepers. Eeueetienel_geele. The decision to keep or release the baby for adoption was related to whether or not the adolescent reported having specific plans to continue beyond her current level of education (Chi-Squarez8.10, df=3, p<.045). All of the "Releasers" had definite educational plans, in contrast to 83.3% of the "Adopting Keepers", 69.2% of the "Aborting Keepers", and 57.1% of the ”Pure Keepers" These data are presented in Table 10. 80 Table 10 Relationship between Outcome and Educational Plans No Specific Specific Row Sim N Juana Elana m 1 Pure Keepers 14 57.1% 42.9% (100.0%) Aborting Keepers 13 69.2% 30.8% (100.0%) Adopting Keepers 14 83.3% 16.7% (100.0%) Releasers 14 100.0% 0.0% (100.0%) Chi-Square=8.10 Df=3 p<.04 Qeegnetien§l_geele. There were no differences between outcome groups with respect to future occupational goals (Chi-Squarez15.19, df=12, p<.23). In five years, nearly one-half (43.4%) of the adolescents would like to hold a professional or executive position; 24.5% would like a technical, sales, or clerical job; and 24.5% expect to be employed in a service job. Only one adolescent (1.9%) reported wanting to work as an operative. Three respondents (5.7%) did not know what job they would like in the future. One adolescent ("Adopting Keeper“) did not answer the question. Furthermore, the adolescent’s perception of her chances of obtaining her ideal job in five years was not related to which pregnancy resolution alternatives she considered (Chi- Square=5.96, df=3, p<.11) or her final choice. Slightly more than one-third (34.0%) of these young women considered 81 that their chances of having the job they would like was "very good", compared to the 66.0% who thought their chances were less good. One respondent (”Aborting Keeper") did not answer. The adolescent’s estimate of her chances of getting the job she wants in five years was unrelated to the type of job she would like to have. flyeetneeje_1e: Releasers will be more likely than keepers to have grown up in a two-parent family. Slightly more than one-third (37.0%) of all adolescents reported living with a single parent some time while they were growing up. Contrary to previous findings in the literature, there was no relationship between number of parents in the home and the adolescent’s decision to keep or release her baby (Chi-Square=.19, df=1, p<.66); forty percent of all adolescents who kept their children and 28.6% of those who chose adoption had only one parent in the home most of the time while they were growing up. Analysis by outcome group showed that 50.0% of the "Pure Keepers”, 42.9% of the "Aborting Keepers", 25.0% of the "Adopting Keepers", and 28.6% of the ”Releasers" spent time growing up in a single-parent home, but group differences were not significant (Chi-Square=2.39, df=3, p<.50). flyeetneeje_g§: Family environment will be related to outcome group. This hypothesis could not be tested. Family environment could not be reliably measured in this sample using Form S of the FES. After obtaining subscales scores, reliability 82 for each of the ten subscales was calculated using Cronbach’s alpha (Norusis, 1988). Coefficients ranged from .25 to .65, falling below accepted standards. Multiple procedures were followed to improve reliability, but none were successful. Content of items was judged for relevance to decision-making, and various items were omitted. Factor analysis of the remaining items was run. No factors with good reliability emerged after several such attempts. Reliabilities were calculated for the three larger dimensions of Relationship, Personal Growth, and System Maintenance, but were also poor. A more detailed review of the procedures followed to establish a reliable measure of family environment is given in Appendix C. H1293h§§i§_£§3 Family environment will be related to the adolescent’s abortion and adoption attitudes, regardless of decision outcome. The relationship of family environment to abortion and adoption attitudes was untestable due to the inability to measure family environment (see Appendix 0). Instead, abortion and adoption attitudes were examined with respect to decision outcome. The findings for abortion were presented above. Respondents’ perceptions that adopted children behave differently from biological children in a family were measured by the scale ”Differ”. Scores ranged from 10 to 20 (with a high score indicating no differences between the two). Cronbach’s alpha (Norusis, 1988) for this five-item 83 scale was .66. There were no significant differences among the outcome groups in their perceptions of adopted children’s behavior (F=.67, df=3, p<.57); the overall mean was 15.9, indicating a fairly strong belief that adopted and nonadopted children do not behave differently from each other. The adolescents’ concern that adopted children will have difficulty being integrated into an adoptive family was related to outcome group (F=3.62, df=3, p<.02). This was measured by the six-item scale "Afraid", with scores ranging from 10 to 24 (with a high score indicating less concern) and a Cronbach’s alpha (Norusis, 1988) of .77 for the scale. Table 11 shows that "Releasers" had the least concern about the child’s integration into the family (mean score of 20.71), in contrast to "Pure Keepers" (mean score of 16.64). Pairwise comparisons using the Scheffe method (Norusis, 1988) indicate that these two groups are significantly different at the .05 level. ”Aborting Keepers" (17.50) and "Adopting Keepers“ (18.00) expressed a similar degree of concern about how adopted children fit into adoptive families. flyeetnee1e_£1: Family environment will be related to the adolescent’s occupational and educational plans for her future, regardless of outcome. This hypothesis could not be tested (see Appendix C). The relationship of outcome group to educational and occupational aspirations was presented above. 84 Table 11 ANOVA of "Afraid" by Outcome Group Sum of Mean Source 12f. Sesame W E 2 Between Groups 3 129.76 43.26 3.62 .02 Within Groups 50 597.57 11.95 Total 53 727.33 QLQEQ N Mean Score * Pure Keepers 14 16.64 Aborting Keepers 14 17.50 Adopting Keepers 12 18.00 Releasers 14 20.71 * "Pure Keepers" and "Releasers" differed at p<.05 using Scheffe pairwise comparisons. No other groups significantly differed from each other. flyeetneeie_£§: Family environment will be related to the adolescent’s perception of influences on her decision. Although this specific hypothesis could not be tested, the adolescent’s perception of influences on her decision was examined with respect to outcome group. This included her perception of the degree to which the choice she made was her own, as well as her perceptions of the ways in which her mother influenced her decision. EeLeentien_efi_eelj_influenee. Outcome groups differed significantly with respect to the degree to which the adolescent perceived her decision to be her own (C.i- Square=16.65, df=6, p<.01). All of the adolescents who considered keeping their babies as their only ootion felt 85 that the decision was "100%” what they wanted to do. However, the adolescents who considered other options reported that their final choices were less than "100%” what they wanted to do to resolve the pregnancy. "Releasers” (35.7%) were least likely to report that their decision was entirely their own. These data are found in Table 12. Table 12 Perception of Own Influence on Decision by Group 100% Mostly 50% Own- Own Own 50% Other Qnenn N Qeeieien Qeeieien Iniinenee Pure Keepers 14 100.0% 0.0% 0.0% Aborting Keepers 14 57.1% 42.9% 0.0% Adopting Keepers 12 66.7% 25.0% 8.3% Releasers 14 35.7% 42.9% 21.4% Chi-Square=10.65 Df=6 p<.01 EeLeeixed_inilnenne_ei_meinenie_e£ien_&e_neln. Eighty- five percent of all adolescents reported mother offered some kind of help for either during the pregnancy or for after the baby was born, and whether or not mother offered help was not related to outcome group (Chi-Square=.89, df=3, p<.83). Nearly two-thirds of the adolescents (60.9%) reported that their decisions would have been more difficult without mother’s offer, and 39.1% reported that the decision 86 was made neither easier or more difficult by mother’s offer to help. Again, the effect of mother’s offer to help on the decision-making process was not related to outcome group (Chi-Square=3.19, df=3, p<.37). Most of the adolescents (88.9%) would have made the same decision even if their mothers had not offered help. Too few adolescents (n=5) reported that their decisions would have been different without their mothers’ help to allow for meaningful statistical analysis. Nonetheless, it is interesting to note that all of these adolescents had at least considered the adoption alternative; three of them were "Adopting Keepers" and 2 were "Releasers". Adolescents in all groups were satisfied with the help that their mothers offered; 94.2% of the sample stated that there was no help that they wanted from their mothers that was not offered. WW There were no differences between outcome groups with respect to how the adolescent’s mother acted toward her during the time she was making her decision (Chi-Square=4.44, df=9, p<.89). Nearly half (46.3%) of the adolescents reported that mother’s actions did not make it either easier or harder for her to do what she wanted about the pregnancy. Forty percent thought mother acted in ways which made it easier for the adolescent to make her own choice. The remaining 13.0% of the adolescents felt that their mothers either worked hard 87 to get them to do what they (mothers) wanted about the pregnancy or made it harder for the adolescents to do what they wanted. Overall, the adolescents’ perceptions were that their mothers did not have a strong influence on their decisions. The degree to which the adolescent felt her mother influenced her decision was not related to outcome group (Chi-Square=9.59, df=6, p<.14). One-half of the respondents (50%) described their mothers’ influence as "not much” or "not at all.” Approximately one-quarter (25.9%) thought mothers influenced their decisions "somewhat", whereas 24.1% reported mothers’ influence as "quite a bit” or "a great deal." This is consistent with the finding that adolescents perceived that their decisions were mostly their own. fixee;hee1e_fifi: Mother’s influence will be related to outcome. Mother’s influence was measured by whether or not mother offered help, and if so, the type of help she offered, as well as by her preference for her daughter’s pregnancy resolution. Of the 85% of the mothers who offered to help their daughters during the pregnancy or afterwards, 95.7% offered practical help and emotional support, as opposed to only emotional support. Examples of practical help included prenatal care, food, clothing, shelter, or financial support. Less than one-quarter (21.7%) of the mothers who offered to help specifically offered help with 88 childcare. No significant differences between groups were found with respect to which mothers offered help, or to the type of help offered. However, mother’s preference for the outcome of her daughter’s pregnancy was related to outcome group (Chi- Square=31.81, df=12, p<.002). As seen in Table 13, half of the ”Pure Keepers" (50.0%) and 41.7% of the keepers who Table 13 Mother’s Outcome Preference by Outcome Group Pure Aborting Adopting Keepenelseenenelseeeeneaeleeeece Mother said keep baby 50.0% 21.4% 41.7% 0.0% Mother said abort baby 14.3% 35.7% 0.0% 0.0% Mother said release baby 0.0% 7.1% 0.0% 28.6% Mother said make own choice 35.7% 28.6% 33.3% 64.3% Don’t know what mother wanted 0.0% 7.1% 25.0% 7.1% Column Total 14.0 14.0 12.0 14.0 Chi-Square=31.81 Df=12 p<.002 89 considered adoption reported that their mothers thought they should keep the baby, compared to about one fifth (21.4%) of the keepers who considered abortion and none of the "Releasers". Compared to any other group, approximately twice as many "Releasers" (64.3%) reported that their mothers encouraged them to make up their own minds about how to resolve the pregnancy. Mother’s outcome preference was not related to whether or not she was working when her daughter became pregnant (Chi-Square=3.66, df=4, p<.46). Neither was her chOice for her daughter influenced by how she felt about the pregnancy (Chi-Square=9.74, df=12, p<.65). H129Lh§§i§_£19= Perception of mother’s influence will be related to mother’s discovery of the pregnancy and living arrangements during the pregnancy. W The timing of mother’s discovery of the pregnancy was not related to the adolescent’s perception of her mother’s influence. All the mothers knew their daughters were pregnant. Thirty-five percent (35.2%) of the mothers reportedly knew of their daughter’s pregnancy before she had a pregnancy test; just under half (46.3%) found out at the time of the pregnancy test; and 19.5% learned about it sometime later. When mother found out about the pregnancy was not related to how she acted toward her daughter while the decision was being made (Chi-Square=3.99, df=6, p<.68) or to the adolescent’s 90 overall perception of her degree of influence on the decision (Chi-Square=7.35, df=4, p<.12). The majority of mothers (72.3%) learned about the pregnancy from the adolescent. WW Nearly two- thirds (63.0%) of the adolescents lived in the parental home while they were pregnant; the remaining 37.0% lived somewhere else. Where the adolescent lived during her pregnancy did not affect how the adolescent perceived her mother to act toward her while she was deciding what to do (Chi-Square=2.56, df=1, p<.11) or perception of mother’s overall influence on the decision (Chi-Square=.60, df=2, p<.75). SUMMARY OF RESULTS Among the adolescent mothers in this study, few differences were found between those who chose to keep their babies and those who released their children for adoption. The mean age of the young mothers was 17 years. Their education was age-appropriate, although more "Releasers” had specific plans to continue their education, compared to the adolescents who kept their babies. About three-fourths had held jobs at some time. The adolescents were raised in nuclear families with an average of one brother and one sister. Just over one third had lived with a single parent at some time. Adolescent pregnancy was not a pattern in the majority of families. 91 Mothers of the adolescents were, on the average, 41 years old. Nearly one-half of the mothers had finished high school, and approximately another third had education beyond high school. Significantly more mothers of "Releasers" were working at the time their daughters got pregnant, but mothers did not differ with respect to the type of jobs they held. This was a first pregnancy for most of the adolescents; nearly all were either not thinking about the possibility of getting pregnant, or they were taking steps not to conceive. Keepers were aware of the pregnancy within the first two months, significantly earlier than releasers, who did not realize it until they were pregnant for nearly three months. ”Pure Keepers" tended to react more positively to the pregnancy, especially compared to "Releasers" and ”Adopting Keepers". All mothers knew of the adolescent’s pregnancy. Most learned of it at least by the time of the pregnancy,test. Most adolescents told their mothers that they were pregnant. Most mothers reacted negatively, but offered practical help as well as emotional support. Nearly two-thirds of the adolescents lived in the parental home during the pregnancy. The adolescents were generally satisfied with the help that _ their mothers offered. Three-quarters of the sample reported that mothers were not a strong influence on the decision to keep or release the baby. However, 92 significantly more mothers of ”Keepers” were reported to have preferred that their daughters keep their babies, compared to mothers of "Releasers", who were more inclined to encourage the adolescents to make their own choices. Compared to adolescents who released their babies, adolescents who kept their babies were more inclined to have perceived their decisions to have been entirely their own. Differences between adolescents who kept or released their babies were found with respect to religious affiliation and attitudes toward pregnancy resolution alternatives. More "Releasers” than ”Keepers” were Catholic, whereas more "Keepers" than "Releasers” tended to be Protestant fundamentalist. The adolescents who kept their babies but had also considered abortion were most liberal in their attitudes toward abortion; they differed significantly from adolescents who kept but also considered adoption as an alternative. With respect to adoption attitudes, all adolescents perceived adopted children as not behaving differently from nonadopted children. However, ”Pure Keepers" were significantly more concerned than "Releasers" about how well an adopted child could be integrated into the adoptive family. Finally, not all of the hypotheses in this study could be analyzed. The measure of family environment had no reliability in this sample. Therefore, the role of family variables and their influence on the adolescent’s decision, 93 both direct and indirect, could not be examined. CHAPTER FIVE DISCUSSION E ] il I . !° One purpose of this study was to describe the individual and family characteristics of adolescents who released their babies and compare them with those of adolescents who kept their children. Overall, the characteristics of the adolescent keepers and releasers in this study followed the changing trends documented in the literature. Adolescents who chose to keep their babies were once‘viewed as socially deviant and psychologically maladjusted. The choice to keep the baby was understood as a reaction to growing up in a broken home or as a result conflictual parental relationships (Vincent, 1960, 1961). Compared to their counterparts who kept their babies, adolescents who chose adoption were described as younger, to be more likely to have come from a two-parent home, to have fewer siblings, to have higher educational and occupational goals, and to come from higher socioeconomic backgrounds. As societal attitudes toward sexual freedom, status of women, divorce, and single-parenthood changed, increasing numbers of adolescents chose to keep their children, and they became the majority among adolescents who continued their pregnancies to term. The concept of social deviance for either outcome decision was refuted (Grow, 1969), but 94 95 the psychological motivation of the adolescent was still considered the basis for her decision. As individual characteristics were examined, differences between keepers and releasers of the late 19603 were less consistently evident, and their meaning became less clear. It was expected that differences between keepers and releasers, if found, would parallel these earlier findings. It is interesting to note, however, that some of the reported characteristics of releasers were quite similar to those of adolescents who opted for abortion; for example, the age for aborters was described as being bimodally distributed, either very young or somewhat older, and adolescents who chose abortion had higher educational and occupational goals compared to keepers (Olson, 1980). Although this study did not address the abortion decision, releaser characteristics paralleling earlier findings were expected since one could posit that adolescents who choose not to parent are similar to each other and different from parenting adolescents along some dimensions. This proved not to be the case. Data from this study showed that adolescents facing pregnancy resolution decision-making about adoption in the mid 19803 were more similar than different, regardless of decision outcome. Keepers and releasers did not differ with respect to a number of individual and family characteristics (e.g. age, educational status, family size, number of parents in the 96 home, future job goals, mother’s age and occupation). However, mothers of releasers were more likely than mothers of keepers to be employed outside the home at the time the adolescent got pregnant, and releasers reported more specific plans to continue their educations in comparison to keepers. Releasers were also further into the pregnancy when they realized they were pregnant, and they reported more negative reactions to the pregnancy. Nonetheless, many of the differences in characteristics which were ambiguous in the literature years ago no longer appear to exist. This lack of differences between keepers and releasers seriously challenges the idea that pregnancy resolution is determined by individual characteristics of the adolescent or the structural aspects of her family. Grow (1979) suggested that the adolescent who chooses to release her child for adoption may hold traditional societal values. Yet apart from religious affiliation, no previous data are available that address values or attitudes among adolescent releasers. Furthermore, no clear relationship exists between decision outcome about adoption and religion (Grow, 1969). In this study, significantly more releasers were Catholic whereas more keepers tended to be Protestant fundamentalist. This finding, unfortunately, is uninterpretable since religion may be confounded with the agency (Catholic Social Services) through which the majority of releasers were recruited. 97 This study also examined the adolescents’ attitudes toward abortion and adoption alternatives. It was expected that no significant differences between groups would be found in abortion attitudes, because none of the adolescents in the sample opted for this choice. However, keepers who also considered abortion were significantly more liberal in their attitudes toward abortion than others, especially the adolescents who also considered adoption. Whereas this may represent a chance finding, it is equally likely that the latter group had reservations about parenting, in that they did consider another alternative, but were opposed to abortion. The reasons why they chose to parent rather than release their babies for adoption were not addressed in this study. More relevant to the adoption decision were the differences found in adoption attitudes, as measured by the scales "Differ" and "Afraid”. None of the adolescents perceived differences between adopted and nonadopted children’s behavior. "Pure Keepers", however, were significantly more concerned than "Releasers” about how an adopted child would be integrated into an adoptive family. This, in part, may explain why adoption was not considered by ”Pure Keepers". If fear of not knowing what will happen to the baby or if s/he will be adequately loved and cared for are significant reasons why more adolescents do not consider the adoption alternative, perhaps more open forms 98 of adoption would promote consideration of this alternative. This position is consistent with findings reported by Barth (1984). More open forms of adoption would allow the adolescent birth mother to have more knowledge about the family with which her child would be placed, and perhaps to maintain some contact over the years. The agency from which most of the releasers for this project were recruited allowed the adolescent to meet three couples and choose the couple she preferred as parents for her child. Although more knowledge of and contact between birth mother and adoptive parents may not be preferable for all adolescents, the impact of more open agency practices on adolescents‘ choices is worthy of further examination. Metneaflnflnenne The second purpose of this study was to determine the adolescent’s mother’s involvement in the pregnancy, as well as her influence on the adoption decision. All the mothers in this study knew of their daughter’s pregnancy. The majority of mothers learned of the pregnancy at least by the time the adolescents got results of the pregnancy test, so the potential to influence the decision began early in the pregnancy. The most common way that mothers learned of the pregnancy was by their daughters telling them. These results are consistent with Rosen (1980), who found that most adolescents did not involve either parent until the pregnancy was confirmed. 99 The majority of mothers of the adolescents in this sample also reportedly offered to help their daughters, even though most mothers reacted negatively to the news. They were perceived by their daughters to act in ways which primarily made the decision either easier or at least not more difficult for the adolescents. Thus, most adolescents did not feel their mothers attempted to gain compliance through use of resources, what Rosen et al. (1982) labeled "direct pressure". The hypothesis that mothers of releasers were not as forthcoming with emotional or practical support in order to influence the adolescent not to keep her child did not prove to be true. It was expected that mothers’ influence on the adoption decision would closely parallel Rosen’s earlier works (1980; Rosen et al., 1982). Interestingly enough, the majority of adolescents, regardless of outcome decision, did not perceive their mothers to have much influence on their decisions. However, mother’s outcome preference was related to final decision: More mothers of keepers wanted their daughters to keep their babies, compared to mothers of releasers, who encouraged their daughters to make their own choices or supported the adoption. Nearly half of the adolescents who considered adoption in addition to keeping reported that their mothers wanted them to keep their babies. It appears that when mother is perceived to want her daughter to keep the child, the child is kept. Contrary 100 to the adolescent’s perception of mother’s influence, but consistent with the literature (Eisen et al., 1983; Hudis & Brazzell, 1981; Musick et al., 1984; Young et al., 1975), mother’s support of the specific decision outcome, rather than her control of resources, appears to be significant in the decision-making process about pregnancy resolution. Perhaps the fact that adolescents do not perceive their mothers as directly influencing their decisions reflects their developmental needs of independence and autonomy. I ! !' E E'Ef In the effort to understand factors relating to the adolescent’s decision-making process, it is easy to presume that individual or family characteristics of the releaser account for her decision to release. This is especially true when differences between keepers and releasers are noted. For example, it could be argued that releasers, who more frequently had role models of mothers working outside the home, were more motivated than keepers to finish their educations and pursue career opportunities; this factor could be considered as one determinant of the decision to choose adoption. However, another equally plausible interpretation of this finding exists. Given the retrospective design of this study, it is not possible to distinguish cause from effect. This point can be illustrated using the same example: It may be that having made the decision to release the baby, 101 the releasers made more specific educational plans in comparison to keepers because they did not have the responsibility of childcare and focused on other goals to help them through their grief. The interpretation that the chosen decision influenced the adolescent’s recall or perception of events having to do with the pregnancy can be explained by cognitive dissonance theory (Festinger, 1957). This theory suggests that whenever a decision is made, postdecisional dissonance arises because there are negative aspects of the chosen alternative and positive aspects of the rejected alternative. To reduce dissonance, which is an aversive state, an individual may alter her cognitions, change her attitudes, or seek social support for the chosen alternative (Festinger, 1957). Any of the differences found between keepers and releasers can be explained from this perspective. This includes not only the adolescent’s educational plans, but also the adolescent’s emotional reaction to the pregnancy, the adolescent’s attitudes toward abortion and adoption, the adolescent’s perception of her mother’s influence and the extent to which the decision was her own, and her perception of mother’s preference for outcome. A prospective design for future research in pregnancy resolution decision-making is recommended to eliminate this difficulty with interpretation. 102 E .1 E . ! The third purpose of this study was to determine how family environment is related to mother’s influence and decision outcome. The family is the social context in which roles and values are learned and attitudes are formed. Characteristics of the family, such as how supportive family members are to each other, whether family members are enmeshed or autonomous, how the family handles problems and conflict, or how organized the family is, were expected to be related to the adolescent’s decision and mother’s influence. These relationships could not be assessed. Surprisingly, the Family Environment Scale (FES) (Form 8) was not a reliable measure of family environment in this sample. The reasons for this finding are speculative. One possible explanation is that this abbreviated form of the FES has poor psychometric properties that are not specific to this group of adolescent mothers. Although poor subscale reliability is not a problem for the 90-item version of the FES (Form R), only one study of reliability was reported for Form S (Moos et al., 1974). Findings from the study were promising, but may not generalize beyond the 11 families in the sample. Given that reliability of a scale is inversely related to the number of items in the scale and to the number of response categories for each item (Kerlinger, 1973), four true-false items may be insufficient to yield adequate subscale reliability for family environment 103 constructs. An alternative, but not mutually exclusive, explanation for poor reliability on the FES (Form 5) is more specific to the use of this scale with a population of adolescent mothers. It is possible that the content of the 40 items was not meaningful from their perspectives. For example, an item such as ”We rarely go to lectures, plays, or concerts" (Moos et al., 1974) may not be relevant to the family life of an adolescent parent. Adolescents who kept their babies are developing new parenting skills, adjusting to changes in the relationship with the baby’s father and other family members, and coping with new responsibilities that include changes in former lifestyle. Releasers, on the other hand, may be grieving the loss of their babies, and their energies may be focused on present and future goals outside the family. With the birth of a baby, the adolescents are in a developmental transition from their families of origin to their families of procreation. However, for adolescent parents, this transition is not normative with respect to expected stages of family life cycle (Carter & McGoldrick, 1980). The transition for adolescent parents is unique in that many live with their parents and rely on them for support, yet they have parenting responsibilities and may maintain some relationship with the baby’s father. They have roles as children and as parents. In this regard, the family life of an adolescent parent is unique. It is 104 possible that Moos’ items (especially in a shorter version of the scale) simply do not capture the family experience that is most relevant for this population of young women. Further, any ambiguity in the perceived meaning of the items would serve to lower reliability since it would add to the error variance of the scale (Kerlinger, 1973). The role of family environment in the decision-making process about adoption is still unclear. It may be that the family variables are related to the adolescent’s values and approach to problem-solving rather than having a direct influence on the decision. In any case; it is also clear that future research in this area needs to expand the definition of family in used in this study to include the baby’s father. WW Although this study did not address decision-making process beyond the relationships of mother’s influence and family environment to outcome, the finding that keepers differ with respect to which, if any, alternatives were considered suggests that the decision-making process may differ among them. Why some adolescents considered other options, as well as which other options they considered, is not clear. It may be that for these adolescents, the proposed models of decision-making (e.g. Bracken et al., 1978; Brazzell & Acock, 1988; Maskay & Juhasz, 1983) are applicable. 105 However, there is other evidence which suggests that the adolescent’s decision to keep her baby is overdetermined (Kallen, Griffore, Popovich, & Powell, 1988). That is, it does not appear to be very subject to influence from others in her social environment, who are perceived by the adolescent as supportive of her decision. From the adolescent’s point of view, it is almost as if there is no choice or decision to be made: She is pregnant, and consequently she will have and raise her baby. Kallen et al. (1988) cited several findings (from the larger data set on which this study is based) which support the overdetermination of the decision among adolescent keepers: (a) Most adolescents who thought about what to do even before getting pregnant planned to keep the baby; (b) once pregnant, the majority adolescents did not consider any other alternatives; (c) keepers made a final decision to keep early in the pregnancy; (d) the adolescents did not welcome the pregnancy, yet chose to keep the baby; (e) any help that was offered by others was not seen as a major influence to keep the baby; and (f) the adolescents saw the choice to keep their babies as their own. Moreover, these findings were identical for both black and white adolescents, supporting a social-structural explanation of adolescent pregnancy. From this perspective, the behavior of pregnant adolescents is scripted, having roots in the social opportunities available to them, in the social and 106 family roles they learn, and in the expectations of others in their environment. Consistent with more recent views, adolescent pregnancy is not thought to be the result of individual psychopathology or a dysfunctional family system. Rather, it is thought to be a result of a series of influences which come from the larger social system. Motherhood is viewed as career, with few negative consequences and a social world in support of the adolescent who begins this career early. There may be ways in which our society benefits from the pregnancies of these young women and supports them in keeping and parenting their children. From a systems perspective, this could be understood as the symptom serving a function. Understanding adolescent pregnancy from this larger societal perspective, as well as understanding the influence of the family as one of the more immediate social contexts of the adolescent, remain areas for future research. I El !' E 1 !° 9 It is apparent from this study that adoption is not currently a viable solution to the problem of adolescent pregnancy. This is evident from the lack of adolescents who consider, let alone choose, adoption as a pregnancy resolution alternative. This study examined approximately equally sized groups of adolescents who considered different alternatives. However, it is important to be mindful of the fact that these groups are not equally distributed in the 107 overall population of pregnant adolescents. Of the 79 white adolescents in the larger data set from which these groups were drawn, 56% (n=44) did not consider any alternative to keeping the baby; 23% (n=18) considered abortion as well as keeping; 15% (n=12) considered but rejected adoption, and 6% (n=5) considered both abortion and adoption. Thus, those who even considered adoption represent only a small fraction of adolescents in the overall study. Furthermore, after two and a half years of arduous field work, only 14 white adolescents who released their babies were recruited. In a broader analysis of differences between keepers and releasers based on the larger study, Kallen, Griffore, Popovich, and Powell (1990) have speculated about which pregnant adolescents may be more likely to consider or choose adoption as an outcome. Results are speculative at best, but it appears that adolescents who are more likely to at least consider adoption, whether or not they actually choose it as an outcome, are those who perceive less social support for keeping their babies, and who also have more definite future plane than adolescents who do not consider adoption. Support from the baby’s father is a key factor, emphasizing the importance of his role in the adolescent’s social world. When adolescents who considered adoption but kept were compared with adolescents who relinquished their children, trends in the data suggest that releasers had less experiences with small children, were less religious, and 108 did not perceive support for keeping the baby from either their mothers or the babies’ fathers. However, these conclusions are based on small samples and remain tentative. Interpretation based on a retrospective design is also problematic. Further research is needed to adequately address the question of which adolescents are likely to choose adoption, and future studies should take into account the influences of the adolescent’s family and the baby’s father in a prospective research design. APPENDICES APPENDIX A 1(39 MICHIGAN ‘STATE UNIVERSITY nomammm wmomemgy "' m "I a telephone“) 517/353-0709 517/353-0883 Dear new bother: We are conducting a study to find out some things about young women who recently had babies. lurley Hedical Center and the flint cmity schools support the study and have given their okay to ask you to take part in it. Bren so, it is up to you to decide whether or not to take part. If you are under the age of 18, we will also need permission tron your parent or guardian. You will receive the sane services tron Hurley hedical Center and the Flint commity schools no natter what your decision about being part of. the study. nyou takepartinthestudy. soueouewilltalkwithyoufor aboutanhourand a half and ask you questions about your experiences, attitudes. and beliefs. The interview will take place in a place convenient for you. You won't have to answer any questions you don’t want to, and you can stop the interview at any tine. arerything you say will be kept confidential. the intonation will be put together with intonation tron other people. and your nane will never be used. Host people we talk to find the experience enjoyable. In order to encourage you to take part, we are offering to give you $10.00 for coepleting the interview. We will also ask you to contact your anther and the baby's father so that we can interview then. We will give you $5.00 for each one that finishes an interview. this neans that you can earn as such as $20.00 if all three of you are interview- ed. In addition. your nother and the baby’s father will each earn $10.00 tor conpleting their interviews. I: you would like nore intonation about the study. you say call the washer“) at the top of the page or Dr. Elizabeth Gordon at 313/257-9516. Sincerely, ‘——-—' _ . c‘ W David .1. lallen. Ph.D. Robert J. Griffore. Ph.D. Professor Chairperson bepartnent of. hdiatrics and We of ranily and linen Dewelopnant Child zoology mama STAT! WWII! HICKS-m STATE WWII! SPIIdn 1 1C) ‘ MICHIGAN STATE UNIVERSITY m U mammal new “IT um 0 m 0 M an ”m ‘Ielephone(s) 517/353-0709 517/353-0333 Dear Parent: You and your daughter have been picked to be part of a study of young anthers and pregnant adolescents., He would like to talk with you about the kinds of help you and your daughter received before her baby was born and about sane of the things you think about for the future. He would also like to know sole of your thoughts and experiences during her pregnancy, as well as how you feel about things-that is. your values and attitudes. Whether you take part in the study is your decision. but Hurley hedical Center and the Flint connunity schools have given us their okay to ask you and your daughter to participate. Your daughter has already agreed that she is willing to be inter- viewed but because she is under the age of 18. we will also need your pernission for her to do so. Your daughter and her baby will receive the sane services free Hurley hedical and the flint connunity schools whether or not you or she take part. If you participate. an interviewer will talk with you for about an hour and ask you a nunber of questions. You won't have to answer any questions you don't want to, and you can stop the interview at any tine. Ewerything you say will he kept con- fidential. Ghat you tell the interviewer will be put together with infornation fron other people so that nothing you say will ever be identified as toning true you. hpst people we talk to find the experience enjoyable. In order to encourage you to take part. we are offering to give you $10.00 for coupleting the interview. If you agree to give your pernission for your daughter to participate, please sign the consent for: on the line above Parent or Guardian. If you would also like to take part in the study, please check the box at the botton of the consent font. we have enclosed a stanped,addressed envelope for you to return the consent fora. If you would like note infornation about the study, you nay call the nunber(s) at the top of the page or Dr. Elizabeth Gordon at 313/257-9516. Sincerely. Zeal... 311 D. lobert .1. Griffore, mp. Professor Chairperson Department of Pediatrics and Deparrnent of Pauly and Susan Developnent Child tcologv EICEIGAS STAIE DRIVERSII! HICEICAR SIAIE UNIVERSITY SPlndn Sign upon contact kespondent ID , szmm mmnmmmmnscnoonsnnmnu 311111111 DEVELOPED?! STUDY cousm FOR}! I have been asked to participate in a study being conducted by David J. Iallen. P‘h.D. and Robert J. Griffore, Pb.D. fron hichigan State University. I have had the procedures fully explained to ne and I have had the opportunity to ask questions. I understand that :7 participation is voluntary and I nay withdraw this consent at any tine. Hy refusal to participate or ey subsequent withdrawal will not affect Iy access to the regular nedical services provided by Hurley hedical Center and the Flint cmunity schools. Turthernore, confidentiality of infor- nation collected has been assured. Sip-astute Social Security masher witness Parent or Guardian Please check the box below if you are willing to be interviewed, (This is for the parent or guardian response.) l___J Please give us your cane, address and telephone umber so we can contact you for the interview (PLEASE PRINT): m: . annazss; SEcial Security mno'er ""’ Telepbone(s): lone ( work ( ) V 1‘12 MICHIGAN ST ATE UNIVERSITY mammals-1m Wm-m-sweuu nut-em ’ Tilephones: $177353-0709 517/353-0883 Dear Friend: as are conducting a study to find out sows things about young wosen who recently had babies. Child and Fasdly Services of Michigan, Inc. supports the study and has given their okay to ask you to take part in it. Even so, it is up to you to decide whether or not to take part. If you are under the age of 18, we will also need per-lesion fro. your parent or guardian. You will receive the sass services frow Child and Easily Services of Hichigan, Inc. no setter what your decision about being part of th- study. If you take part in the study, soseone will talk.with you for about an hour and a half and ask you questions about your experiences, attitudes, and beliefs. The interview will take place in a place convenient for you. You won‘t have to answer any questions you don't want to, and you can stop the interivew at any ties. Everything you say will be kept confidential. The inforeation will be put together with inforwation fros other people, and your nape will never be used. Host people we talk to find the experience enjoyable. In order. to encourage you to take part, we are offering to give you 310.00 for coupleting the interview. We will also ask you to contact your sother and the baby‘s father so that we can interview then. We will give you $3.00 for each one that finishes an interview. This seans that you can earn as such as $20.00 if all three of you are interviewed. In addition, your anther and the baby's father will each earn 310.00 for coupleting their interviews. .If you would like sore inforsation about the study, you say call the nusber(s) at the top of the page. We would be happy to answer any questions you eight have. If you would like to be interviewed, please fill out the consent fore and return it to us in the steeped, addressed envelope that is enclosed. Sincerely, L /.‘ David J. Kellen, Ph.D. Robert .1. Griffore, Ph.D. Professor Chairperson and Professor Depart-eat of Pediatrics and Depart-eat of Fasily and Husan Develop-sat Child Ecology HICHISAM STATE UNIVERSITY HICHISAN STATE UNIVERSITY .m’banHWln-wns 1‘13 MICHIGAN STATE UNIVERSITY nomaindotnmunmmumuwnumwnnnr murumumzomounu-aouni I” an m Telepboeu) 517/353-0709 Dear Parents 517,353.03” . You and your daughter have been picked to be part of a study of young bothers and pregnant adolescents. He would like to talk with you about the kinds of help you and your daughter received before her baby was born, about her decision to release the baby for adoption, and about son of the things you think about for the future. we would also like to know sons of your thoughts and experiences during her pregnancy, as well as how you feel about things-thet is, your values and attitudes. whether you take part in the study is your decision, but Child and Faeily Services of hichigin, Inc. has given us their okay to ask you and your daughter to participate. You daughter has also received a letter, but because she is under the age of 18, we will need your pereission as well if she would like to participate. She will receive the same services froe Child and Faeily Services of Hichigen, Inc. whether or not you or she take part. If you participate, an interviewer will talk with you for about an hour and ask you a nueber of questions. You won't have to answer any questions you don't want to, and you can stop the interview at any tine. Everything you say will be kept confidential. Hhet you tell the interivewer will be put together with inforeetion froe other people so that nothing you say will ever be identified as coeing froe you. This all applies to your daughter's interview as well. Host people we talk to find the experience enjoyable. In order to encourage you to take part, we are offering to give you $10.00 for coepleting the interview. If you agree to give your pereission for your daughter to participate, please sign the consent fore on the line above 'Parent or Guardian's Signature“. If you would also like to take part in the study, please cooplete the Inforsetion Sheet attached to this letter. He have enclosed a steeped, addressed envelope for you to return the Consent Fore and Inforeetion Sheet. If you would like sore inforeation about the study, you say call the nueber(s) at the top of the page. He would be happy to answer any questions. Sincerely, a— ) m/ l/ . David J. Kellen, Ph.D. Robert J. Griffore, Ph.D. Professor ‘ Chairperson and Professor Department of Pediatrics and Deperteent of Faeily and Hueen Develop-ent Child Ecology HICHIBQN STATE UNIVERSITY MICHIGAN STATE UNIVERSITY ...-__ .. ....,.. I....IWW 114 INFDRHATION SHEET PLEASE CHECK ONE: NO: I would not like to be interviewed as part of the study, but I do give ey pereission for by daughter to do so. I understand that ey daughter will receive the sane services froe Child and Faoily Services of Hichigan, Inc. regardless of oy decision. YES: I would like to be interviewed and receive £10.00 for completing the interview. If you answered YES, please fill in the following inforeation so that we can contact you: NAHE: ADDRESS: APPENDIX B APPENDIX 8 Following are the actual interview questions and response or coding categories for the variables used in this study. The questions are a subset of the interview schedule used for the overall study. They are organized here by the variables being considered in this study and not in the order of presentation to the subject. The variables are described in Chapter 3. If the question was a close-ended question, the responses appear just after the question. If “coding category" appears after the question, the question was open-ended and later coded according to the categories listed. The interviews were coded by trained coders who reached an intercoder reliability coefficient of .80 or above. W Adolescent’s age QUESTION: How old are you now? CODING CATEGORIES: Actual age of respondent in years Education QUESTION: How far have you gone in school? CODING CATEGORIES: (preceded by interviewers) last grade completed 98=trade school 115 116 Religion QUESTION: What religion are you? CODING CATEGORIES: 1=Protestant, liberal 5=Jewish 2=Protestant, fundamentalist 6=Moslem 3=Mormon 7=None / atheist 4=(Roman) Catholic 8=Other Employment QUESTION: Another thing we are interested in is your experience in working. Have you ever held a job? 1=Yes 2=No Pregnancy History QUESTION: Was your pregnancy with (name of child) your first pregnancy or not? 1=Yes 2=No QUESTION: (If No) How many times were you pregnant before your pregnancy with (name of child)? CODING CATEGORIES: number of previous pregnancies QUESTION: (If more than one pregnancy) What happened as a result of your earlier pregnancies? Did you: Have a live baby and keep it? 1=Yes 2=No Have a live baby and release it for adoption? 1=Yes 2=No Have an abortion? 1=Yes 2=No Have a miscarriage? 1=Yes 2=No Something else (what)? 117 Abortion Attitudes These questions are taken from Premarital Abortion Permissiveness (PAS) scale by Mirande and Hammer (1974). QUESTIONS: We are also interested in your opinions about when, if ever, it is all right for a woman to have an abortion. The answers this time go from strong, medium, or slight agreement to slight, medium, or strong disagreement. * It is mayo; all right for a woman to have an abortion for any reason. It jo all EIQDL foc a woman to h§¥§ an EDQELIQD1 If a doctor told her it was necessary in order to prevent serious injury. If she felt she could not provide the kind of life she wanted her children to have. If she felt being pregnant would make it difficult for her to continue her education. If she would have to give the child up for adoption. SCORING: An asterisk (*) before the item indicates that the item was reverse scored. Scores range from 5 to 30 points. A high score indicates opposition to abortion, while a low score suggests a more permissive attitude toward abortion. Adoption Attitudes XABIABLE: Differ QUESTIONS: Next are some statements that people sometimes make about adopted children. For each one, please tell me whether you strongly agree, agree, disagree or strongly disagree. * Adopted children love their adoptive parents as much as natural children love their parents. Adoptive parents can expect to be less satisfied as parents than couples who have their own children. * Adopted children are just like other children. 118 Adopted children are more likely to get into trouble than other children. Adopted children can never really be happy because they will always wonder who their real parents are. SCORING: An asterisk (*) before the item indicates that the item was reverse scored. Scores range from 5 to 20, with a high score indicating that adopted and nonadopted children do not have major differences. XABIABLB Afraid QUESTIONS: Now I’m going to read you a list of fears that some couples have when they think about adopting a child. For each, I’d like you tell me whether you think it is something they should be very afraid of, afraid of, not so afraid of, or not at all afraid of: Afraid of telling the child that he or she is adopted. Afraid they won’t love the child like their own. Afraid the child will never accept them as his or her real parents. Afraid their family won’t accept the child. Afraid they won’t be good parents to an adopted child. Afraid the child will always wonder who his or her real parents are. SCORING: Scores range from 6 to 24. A high score indicates the adolescent is not worried about how an adopted child adjusts within the adoptive family, while a low score suggests she has concerns. Future Goals XABIABLE: Education QUESTION: What are you doing about school now? (Coded by interviewer without reading categories) 1=Reports specific plans for continuing school 2=Has general intentions, but no specific plan 3=Confused 4=No plan 119 XABIABLE: Occupation QUESTION: (Think ahead to the future for a minute. Think ahead to the time when your baby has grown up a bit and is in school.) Ideally, what kind of job would you like to have then? CODING CATEGORIES: Occupation was coded according to the Classified Index of Industries and Occupations (U.S. Department of Commerce, 1980). QUESTION: How good do you think your chances will be of getting a job like that in the future? Would you say: i=Very good 2=Fairly good 3=All right 4=Not so good 5=Not good at all W Socioeconomic Status XABLABLE: Mother’s education QUESTION: And how far in school did she (mother) go? CODING CATEGORIES: 01-11 = Number corresponding to last grade completed 12 = High school graduate or has a GED 12 + x = Where x is years of college or junior college 25 = Went to trade school 99 = Don’t know XABLABLE: Mother’s employment and occupation QUESTION: Did she have a job at the time you got pregnant? 1=Yes 2=No 120 QUESTION: What did she do on her job? CODING CATEGORIES: Occupation was coded according to the Classified Index of Industries and Occupations (U.S. Department of Commerce, 1980). Family Structure XABLABLE: Mother’s age QUESTION: How old is your mother now? CODING CATEGORIES: Mother’s age in years XABLAQLE: Family history of adolescent pregnancy QUESTION: Were any members of your household involved in a pregnancy themselves before they finished high school? 1=Yes 2=No XABLAELE: Number of parents QUESTION: Were both your (biological, foster, adoptive, etc) parents living at home most of the time while you were growing up? 1=Yes 2=No XABLAELE: Household composition QUESTION: Who else, if anyone, was living in your home most of the time while you were growing up? CODING CATEGORIES: Number of sisters ("real", half, and step) Number of brothers ("real", half, and step) Number of other relatives (extended family members) 121 Family Environment The Family Environment Scale-Form S was reproduced by special permission of the Publisher, Consulting Psychologists Press, Inc., Palo Alto, CA from Eamj]y Enligonmon1_§oalo by Moos et al. (1974). Further reproduction is prohibited without the Publisher’s consent. The FES-Form S is divided into ten subscales, four items each. The items are listed below according to subscale, but this is not the order of presentation to the respondent. The respondent was read the general directions and then the items. All items were judged to be TRUE or FALSE, and scoring was done according to the manual (Moos et al., 1974). (T) or (F) after the question refers to the answer that is scored as a point on the subscale. Subscale raw scores range from O - 4. Standardized subscale scores will be classified into typologies according to Moos and Moss (1976). GENERAL DIRECTIONS: I’m going to read some statements about families, and I’d like you to tell me whether each statement is true or false for the family that you grew up with. The statement may be true for some members and false for others. In that case, tell me whether it is true or false more moo; members. SUBSCALE: Cohesion QUESTIONS: Family members really help and support one another. (T) We often seem to be killing time at home. (F) We put a lot of energy into what we do at home. (T) There is a feeling of togetherness in our family. (T) SUBSCALE: Expressiveness QUESTIONS: Family members often keep their feelings to themselves. (F) We say anything we want to around home. (T) It’s hard to "blow off steam” at home without upsetting somebody. (F) We tell each other about our personal problems. (T) 122 SUBSCALE: Conflict QUESTIONS: We fight a lot in our family. (T) Family members rarely become openly angry. (F) Family members sometimes get so angry they throw things. (T) Family members hardly ever lose their tempers. (F) SUBSCALE: Independence QUESTIONS: We don’t do things on our own very often in our family. (F) In our family, we are strongly encouraged to be independent. (T) We think things out for ourselves in our family. (T) We come and go as we want to in our family. (T) SUBSCALE: Achievement Orientation QUESTIONS: We feel it is important to be the best at whatever you Gggtin;)ahead in life is very important in our family. How)much money a person makes is not important to us. W%:§elieve in competition and "may the best man win“. SUBSCALE: Intellectual-Cultural Orientation QUESTIONS: We often talk about political and social problems. (T) We rarely go to lectures, plays or concerts. (F) Learning about new and different things is important to us. (T) We are not that interested in cultural activities. (F) 123 SUBSCALE: Active-Recreational Orientation QUESTIONS: We spend most weekends and evenings at home. (F) Friends often come over for dinner or to visit. (T) Nobody in our family is active in sports, Little League, bowling, etc. (F) We often go to movies, sports events, etc. (T) SUBSCALE: Moral-Religious Emphasis QUESTIONS: Family members attend church, synagogue, or Sunday School fairly often. (T) We don’t say prayers in our family. (F) We often talk about the religious meaning of Christmas, Passover, or other holidays. (T) We don’t believe in heaven or hell. (F) SUBSCALE: Organization QUESTIONS: Activities in our family are pretty carefully planned. (T) We are generally very neat and orderly. (T) It’s often hard to find things when you need them in our household. (F) Being on time is very important in our family. (T) SUBSCALE: Control QUESTIONS: Family members are rarely ordered around. (F) There are very few rules to follow in our family. (F) There is one family member who makes most of our decisions. (T) There are set ways of doing things at home. (T) 124 EBEGUANQI:BELAIED_¥ABIAELE§ Pregnancy Intentions QUESTION: At the time you got pregnant with (name of child), would you say you were: 1=Trying to get pregnant 2=Not thinking about the possibility of getting pregnant 3=Trying to keep from getting pregnant Self Discovery of the Pregnancy MABIABLE= QUESTION: XABIABLE: QUESTION: When the adolescent realized she was pregnant Now I’d like to ask you about your pregnancy. To begin with, how far along were you when you first thought you might be pregnant? I CODING CATEGORIES: Actual number of months into pregnancy Adolescent’s reaction How did you feel when you found out that you were pregnant? CODING CATEGORIES: 1=Positive feelings 2=Negative feelings 3=Confused or mixed feelings 4=Surprise or shock, no other emotion 5=Not surprised or shocked, no other emotion 6=Shock or surprise and positive emotions 7=Shock or surprise and negative emotions 8=Other,(what?) 125 Mother’s Discovery of the Pregnancy XABIAELE: QUESTION: XABIABLE: QUESTION: MABIABLE= QUESTION: When mother found out When did your mother first know that you were pregnant? Would you say that it was: 1=Before you even thought you might be pregnant 2=When you first thought you might be pregnant 3=When you first got the results of the test 4=Later in the pregnancy 5=Not until after the baby was born 6=Never How mother found out Which of the following best describes how your mother found out that you were pregnant? 1=She asked you and you told her 2=You told her without her asking 3=You were showing so she could see you were pregnant (physical or behavioral changes) 4=Someone also told her 5=She found out some other way (How?) How did she feel when she found out? How did she feel when she first found out? CODING CATEGORIES: 0=Don’t know 1=Positive feelings 2=Negative feelings 3=Confused or mixed feelings 4=Surprise or shock, no other emotion 5=Not surprised or shocked, no other emotion 6=Shock or surprise and positive emotions 7=Shock or surprise and negative emotions 8=Other (what?) 126 Living Arrangements During Pregnancy QUESTION: Where were you living most of the time during your pregnancy? (Coded by interviewer without reading responses) 1=In parental home 2=In own home by self 3=In own home with father of baby 4=In own home with friend 5=In home Of a relative 6=In home Of baby’s father (with his family) 7=Other (where?) Self Influence on the Decision ‘1A31ABLE: Alternatives considered Adolescents were all asked whether they considered each of the three alternatives (abortion, adoption, keeping) when they were thinking about what to do about their pregnancies. Respondents were then categorized by outcome and which alternatives they considered. This variable is used as a subject variable to increase sample size and to determine differences and similarities that may be related to alternatives considered but not outcome. The actual questions are given below. QUESTIONS: (For keepers) When you were thinking about what to do about your pregnancy, would you say that having an abortion was something that you thought about doing or something that you could never do? 1=Thought about doing 2=Could never do And would you say that giving your baby up for adoption was something that you thought about doing or something that you could never do? 1=Thought about doing 2=Could never do 127 (For releasers) When you were thinking about what to do about your pregnancy, would you say that having an abortion was something that you thought about doing or something that you could never do? 1=Thought about doing 2=Could never do And would you that keeping your baby and raising it was something that you thought about doing or something that you could never do? 1=Thought about doing 2=Could never do Mother’s Influence on the Decision MABIABLE: QUESTION: lABIAflLEi QUESTION: QUESTION: Mother’s preference for outcome When your mother first knew about your pregnancy, what did she think you should do about it? 1=Have the baby and keep it 2=Have an abortion 3=Have the baby and give it up for adoption 4=Have the baby and let relatives raise it 5=Make your own choice 6=She didn’t care 7=She never said what she thought you should do 8=Something else (what?) Resources offered Did your mother ever offer you any help either for while you were pregnant or later? 1=Yes 2=No (If yes) What kind of help did she Offer? CODING CATEGORIES: 1=Emotional support only, no material help 2=Material or practical support (with or without mention of emotional support) 3=Other (what?) WEI-E: QUESTION: QUESTION: Perception QUESTION: Perception W: QUESTION: 128 Mother’s help was also coded as to whether or not it involved help with child care: 1=Yes 2=No Resources not offered Was there some help that you wanted from your mother that she did not give? 1=Yes 2=No (If yes) What help was that? CODING CATEGORIES: 1=Emotional support only, no material help 2=Material or practical support (with or without mention of emotional support) 3=Other (what?) of Own Influence In deciding what to do, which Of the following best describes how you made your choice? Would you say: 1=It was 100% what you wanted to do 2=It was mostly what you wanted to do 3=It was about half what you wanted to do and about half what other people wanted you to do 4=It was mostly what other people wanted you to do 5=It was 100% what other people wanted you to do of Mother’s Influence Effect of mother’s help Would your decision about what to do have been easier, more difficult, or no different if this help had not been Offered? 1=Easier 2=More difficult 3=No different QUESTION: QUESTION: XARIAELE: QUESTION: QUESTION: QUESTION: 129 DO you think you would have made a different decision if this help had not been offered? 1=Yes 2=No (If yes) How might it have been different? CODING CATEGORIES: 1=Would have had an abortion 2=Would have released baby 3=Other 4=Would have kept baby Effect of lack of mother’s help (If mother did not offer help adolescent wanted) Would your decision about what to do have been easier, more difficult, or not different if this help had been Offered? 1=Easier 2=More difficult 3=No different (If mother did not offer help adolescent wanted) Do you think that you would have made a different decision if this help had been Offered? 1=Yes 2=No (If Yes) How might it have been different? CODING CATEGORIES: 1=Would have had an abortion 2=Would have released baby 3=Other 4=Would have kept baby ¥ABIABLE= QUESTION: XABIABLE: QUESTION: Outcome XABIAELEI 130 Mother’s actions All things considered, which would you say best describes how your mother acted towards you while you were deciding what to do about your pregnancy? Would you say: 1=She worked hard to get you to do what she wanted 2=She made it easier for you to do what you wanted 3=She made it harder for you to do what you wanted 4=She really didn’t make it easier or harder for you to do what you wanted Degree of influence And how much would you say your mother influenced your decision? Would you say: 1=She 2=She 3=She 4=She 5=She influenced you a great deal influenced you quite a bit influenced you somewhat didn’t influence you so much didn’t influence you at all Outcome group This is determined by the adolescent’s final choice and the alternatives she considered (see Self Influence, above). The four groups are "Pure Keepers“, "Aborting Keepers", ”Adopting Keepers", and "Releasers". APPENDIX C APPENDIX C The FES (Form S) was not a reliable measure of family environment in this sample of young adolescent mothers. Several steps were followed in an attempt to improve reliability and to create a reliable measure of family environment using the data from this scale for this study. None Of these attempts were successful. This appendix describes the procedures that were followed before the decision was reached that family environment could not be measured by this data set. We The forty items of the FES (Form 8) were scored according to manual directions (Moos et al., 1974) to generate scores for the ten subscales. Reliability of each subscale was then calculated using Cronbach’s alpha (Norusis 1988). As can be seen from the data in Table C.1, the reliabilities Of most subscales fall far below acceptable standards. It should be noted that deletion of particular items from the subscales would not sufficiently increase the alpha coefficients. With reliability coefficients so low, any additional analysis using subscales (e.g. cluster analysis of subscales to create typologies) was meaningless. 131 132 m Next, an attempt was made to extract different factors or subscales from the 40 items that would hopefully have better reliability in this sample. Factor analysis yielded Table 14 Reliabilities for the FES (Form 8) Subscales Scale Standard Cronbach’s Subscale Mean Deflation _ILQA h Cohesion 2.907 0.937 .3799 Expressiveness 2.074 1.286 .6053 Conflict 2.074 1.096 .4473 Independence 2.982 0.921 .3096 Achievement Orientation 2.537 0.863 .2543 Intellectual-Cultural Orientation 2.019 1.107 .5084 Active Recreational 2.943 1.045 .4560 Moral-Religious Emphasis 2.537 1.255 .6505 Organization 2.667 1.082 .3927 Control 2.352 1.168 .4541 15 factors. The major problem with the results Of this factor analysis was that varimax rotation was unsuccessful after 24 iterations. The results were uninterpretable, since most factors in an unrotated factor matrix correlate with a number variables (Norusis, 1988). Further, examination of the interitem correlation matrix generated by this factor analysis leads one to question whether factor analysis was an appropriate statistical procedure for this data in the first place. Factor analysis assumes there is some relatedness between items that can be accounted for by 133 underlying factors (Norusis, 1988). The correlation matrix from this analysis, however, showed very low interitem correlations indicating little relationship between the 40 items in this scale. SLSD_Ih£9§ Another strategy was to delete various items from the 40 item pool and to rerun factor analysis on the item subset. Items were deleted based on content; items judged to be less relevant to decision making around pregnancy were dropped (e.g. Nobody in our family is active in sports, Little League, bowling, etc.). It was hoped that this would improve correlations between remaining items. Although this strategy was attempted several times, none of these attempts were successful in increasing interitem correlations or in producing independent factors that were based on more than two items. W Next, individual items were grouped according to the dimensions outlined by Moos et al. (1974). Reliability coefficients were calculated for the dimensions of Relationship, Personal Growth, and System Maintenance (Moos et al., 1974). It was hoped that combining more items in the unit of analysis would increase reliability coefficients. Cronbach’s alpha (Norusis, 1988) was calculated for each dimension. The reliability coefficient for the Relationship Dimension was .1977; Personal Growth 134 was .5292, and the System Maintenance coefficient was .5525. Although the latter two may reflect an improvement in magnitude over the reliability coefficients of most subscales, all Of these values were too far below accepted reliability standards. In any case, interpretation Of these dimensions would have been problematic. SIELEJ'JLE Given that analyses based on individual items had proven totally unsuccessful, analysis at the subscale level was continued. Although the subscales were meaningless because of poor reliability, a second order analysis of them was attempted. Factor analysis of subscale scores was done on the sample for this study and on the larger sample of white keepers. These two analyses produced three and four factors, respectively, with different subscales loading on the factors within each sample. This discrepancy in findings indicated that the factors extracted in these analyses were also unreliable, because 40 adolescents in the smaller sample (n=54) were drawn from the larger one. Conclusions The FES (Form 8) was not a reliable measure Of family environment in this sample. Subscales had poor reliability and items were not closely related. Analyses following Moos’ conceptualizations of dimensions (Moos et al., 1974), as well as other attempts to creatively manipulate the data to derive some measure of family environment were all 135 unsuccessful. Possible reasons for these results are discussed in Chapter 5. REFERENCES REFERENCES Alan Guttmacher Institute. (1981). Ioonaoo oroonancy; Tho Q_QDl§m_Ln§&_n§§n_&_QQn§_fiflax- New York: Alan Guttmacher Institute. ’ Ascher, C. (1985). Pregnant and parenting teens: - statistics characteristics, and school-based support 3 services. ERIC/CUE Trends and Issues Series, Number 1. ’ Bachrach, C.A. (1983). Adoption as a means of family formation: Data from the national survey of family growth. Joornal Of Macriago ano the Family, ii, 859-865. Bachrach, C.A. (1986). Adoption plans, adopted children, and adoptive mothers. rn l f M rri n th Family, go, 243-253. Barth, R.P. (1987). Adolescent mothers’ beliefs about open adoption. §ooi a] anonork (June), 323-331. Barth, R.P., Schinke, S.P., & Maxwell, J.S. (1983). Psychological correlates of teenage motherhood. Journal W. 1.2.. 471-487. Bracken, M. (1973). Abortion attitudes and discrepant behavior: Some theoretical issues. Hoalth Eoucation Journal, £2: 4-9. Bracken, M., Hachamovitch, M. & Grossman, A. (1974). The decision to abort and psychological sequelae. ‘ l f N rv n r s, 15,155-151. Bracken, M., & Kasl, S. (1975a). Delay in seeking induced .abortion: A review and theoretical analysis. Amerjoan Joornal of ODSLELFIQ§ ano Gynooology, 12], 1008-1019. Bracken, M., & Kasl, S. (1975b). First and repeat abortions: A study of decision- making and delay. Joornal of Ejooooi al §oi onoo, 473- 491. 136 137 Bracken, M., Klerman, L., & Bracken, M. (1978). Abortion, adoption, and motherhood: An empirical study of decision- making during pregnancy. Amacjoan JQHEDE] of W. 1.3.0 251- 262 Brazzell, J.F., & Acock, A.C. (1988). Influence of attitudes, significant others, and aspirations of how adolescent intend to resolve a premarital pregnancy. W41 59. 413- 425 Busch-Rossnagel, N.A. (1985). Review of Family Environment Scale. In J.V. Mitchell, Jr. (Ed.), Nintn Manta] Meaooromento yoacoook (Vol. 1, pp. 573-574). Lincoln, NE: University of Nebraska Press. Butts, R.Y., & Sporakowski, J.J. (1974). Unwed pregnancy decisions: Some background factors. Joocnal of Soy Roooaron, 19, 110-117. Carter, E.A., & McGOldrick, M. (Eds.) (1980). The family lifogcyclo: Aoframonock for family LDECQQX- New York: Gardener Press, Inc. Chilman, 0.5. (1978). l n l' in n 'n Am ri O ' - ' l h l i r c*iv Washington, 0.0.: U.S. Department of Health, Education, & Welfare. Chilman, 0.5. (1979). Teenage pregnancy: A research review. §ooial Work, 25, 492-496. Clapp, D.F., & Raab, R.S. (1978). Follow-up of unmarried adolescent mothers. Soojal Work, 23, 149-153. Eisen, M., & Zellman, G.L. (1984). Factors predicting pregnancy resolution decision satisfaction of unmarried adolescents. Tho Joanna] of QSDELlQ Poyohology, 14; 231-239. Eisen, M., Zellman, G.L., Leibowitz, A., Chow, W. K. & Evans, J. R. (1983). Factors discriminating pregnancy resolution decisions of unmarried adolescents. G n i Poyohology Monogr aoho, _oa, 69- -95. Elder, G.H., Caspi, A., & Burton, L.M. (1988). Adolescent transitions in developmental perspective: Sociological and historical insights. In M. R. Gunnar and W. A. Collins’ (Eds. ) Qoyoloom man 3 ouring tho tcan§1ti on to aooleocenoe; minnooota oynoooja on ohilo osychology (Vol. 21) (00. 151-179). Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. 138 Evans, J., Selstad, G. & Welcher, W. (1976). Teenagers: Fertility control behavior and attitudes. Eanily W 8. 192- 200. Festinger. L. (1957). Wage Evanston, IL: Row, Peterson. Festinger, T.B. (1971). Unwed mothers and their decisions to keep or surrender children. gnjlo_uoliano, an, 253-263. Fischman, S. (1975). The pregnancy resolution decision of unwed adolescents. Ina Nonajng glinjo of NQELD mm. .LQ. 217-227 Friedman, H.L. (1975). Why are they keeping their babies? §oojal york, 29, 322-323. Furstenberg, F. F. Jr. (1981). Implicating the family: Teenage parenthood and kinship involvement. In T. Ooms (Ed ). IEEnaaa_pLEaaansx_in_a_£amilx_sgntaat (DD. 131— 164). Philadelphia: Temple University Press. Furstenberg, F.F., Jr., & Brooks-Gunn J. (1985). Teenage childbearing: Causes, consequences, and remedies. In L. H. Aiken and D. Mechanic (Eds. ), Aoo]joatjon§ of aocja] 334). New Brunswick, NJ: Rutgers University Press. Furstenberg, F.F., Jr., Brooks-Gunn, J., S Morgan, 8. P. (1987). W. New York: Cambridge University Press. Furstenberg, F. F. , Jr. & Crawford, A. G. (1978). Family support: Helping teenage mothers to cope. Family Elfiflfllflfl_E§£§E§§Ll¥§§ lQ: 322' 333 Furstenberg, F.F., Jr., Morgan, 8. P. Moore, K. A. & Peterson, J. L. (1987). Race differences in the timing of adolescent intercourse. Amonloan_§oo1olooical Review, 52, 511- -518. Grow, L. J. (1969L The unwed mother who keeps her child. In The goonl o joooacoy- Ina tcjol a or jaja; Ill og1ti maoy LQQQX (pp. 115- 125). New York: National Council on Illegitimacy. Grow, L.J. (1979). Today’s unmarried mothers: The choices have changed. anlo nolfaca, £8. 363-371. Grow, L.J., & Smith, M.J. (1971). Adoption trends: 1969- 1970. Child Holfaco, an, 401-408. 139 Haring, B. (1975). Adoption trends, 1971-1974. Child flfiliéifi. 5A. 524-525- Hass, P. H. (1974). Wanted and unwanted pregnancies: A fertility decision- making model. Jogcnal of Sooja] ISSUES 39. 125- 165. Henshaw, S.K. Binkin, N. J. , Blaine, E., a Smith, J. C. (1985). A portrait of American women who Obtain abort1ons.Eam111_Elannlns_EEL§2E23.1aa..11. 90 95- Henshaw, S.K., & Van Vort, J. (1989). Teenage abortion, birth, and pregnancy statistics: An update. Family W. 2.1. 85-88. Hudis, P.M., & Brazzell, J.F. (1981). Significant others, adult-role expectations, and the resolutions Of teenage pregnancies. In P. Ahmed (Ed.), Pcognanoy, onilooirth, ano_oaLontnooo (pp. 167-188). New York: Elsevier. Jones, W.C., Meyer, H.J., & Borgatta, E.F. (1962). Social and psychological factors in status decisions of unmarried mothers. f ' d th F .il , 24, 224-230. Kallen, D.J., Griffore, R., Popovich, 8., & Powell, v. (1990, June). Some datorminants of [alinquishment r n l c n . Paper presented at the Conference on Adoption Research sponsored by the Office of Adolescent Pregnancy Programs, Washington, DC. Kallen, D.J., Griffore, R., Popovich, 5., & Powell, V. (1988). Adoption decisions: Personal and social context. Final Report of Project APR 000915, Office of Adolescent Pregnancy Programs, Department of Health and Human Services. Kallen, D. J. , Popovich, S., Griffore, R. & Powell, V. (1988, November). Adolosoentm moahera ano Lnoir mothers viaw aoootjon. Paper presented at the National Council on Family Relations, Philadelphia, PA. Kallen, D.J., & Stephenson, J.J. (1982). Talking about sex revisited. Joanna! of yootn and Aooloaoanoa, 11, 11-23. Kerlinger, F.N. (1973). Foondationa of behavioral raaearch (2nd edition). New York: Holt, Rinehart, and Winston, Inc. Kirk, H.D. (1964). Sharon fata; A theory of adootion and mantal nealtn. Glencoe, IL: The Free Press. 140 Leibowitz, A., Eisen, M., & Chow, W. (1980). - i i n kin in r a - l i 1 Santa Monica, CA: The Rand Corporation. Lewis, 0.0. (1980L A comparison of minors’ and adults’ pregnancy decisions. Amocjoan Joannal of W. 5.9. 446-453. Leynes, C. (1980). Keep or adopt: A study of factors influencing pregnant adolescents’ plan for their babies. WW. .1_. 105-112. Maskay, M.H., & Juhasz, A. M. (1983). The decision- -making process model: Design and use of adolescent sexual decisions. Eamjly Bolatjona, 32, 111- -115. McLaughlin, 8.0., Grady, W.R., Billy, J.O., Landale, N.S., & Winges,L.D. (1986). The effects of the sequencing of marriage and first birth during adolescent. Family Planning E§E§D§Q§jV§§. 15, 12-18. Mecklenburg, M.E., & Thompson, P.G. (1983). The adolescent family life program as a prevention measure. Poolio was. 9.8. 21-29. Meyer, H.J., Borgatta, E.F., & Fanshel, O. (1959). Unwed mothers’ decisions about their babies: An interim replication study. anlo Wolfaca, go, 1-6. Meyer, H. J. , Jones, W., & Borgatta, E. F. (1955). The decision by unmarried mothers to keep or surrender the r babies. §oojal flock, l, 103- 109. Mirande, A.M., & Hammer, E.L. (1974). Premarital sexual permissiveness and abortion. Paoifjo Sooiologjoal 3:11:21. 1 , 485-503. Mitchell, J.V., Jr. (Ed.). (1985). Ninth Mental M a m t Y r 0 (Vol. 1). Lincoln, NE: University Of Nebraska Press. Moore, K.A., Peterson, J.L., & Furstenberg, F.F. (1986). Parental attitudes and the occurrence of early sexual activity. Joanna] of Marciano and the Family, aa, 777 ~782. Moore, K.A., Simms, M.C., & Betsey, C.L. (1986). Choice n i c m ° a i l i e n s in a 01 ce * se ua ' n 'li . New Brunswick, N.J.: Transaction Books. 141 Moos, R. H. Insel, P.M., & Humphrey, B. (1974). W1 for familLW WWW- Palo Alto, CA: Consulting Psychologists Press, Inc. Moos, R.H., & Moos, B.S. (1976L A typology of family social environments. Eamjly_ELooaa§, _5, 357-371. Musick, J.S., Handler, A., & Waddill, K.D. (1984). Teens and adoption: A pregnancy resolution alternative? W. 1.3. 24-29. National Research Council. (1987). Bisking tha fotura: Aooloaoont aoxoality, ocagnanoy, and ohildbaaring (Vol. 1). Washington, 0.0.: National Academy Press. Norusis, M.J. (1988). P P + n d ta i ics V2,Q. Chicago: SPSS, Inc. O’Connell, M., & Rogers, 0.0. (1984). Out-Of—wedlock births: Premarital pregnancies and their effect on family formation and dissolution. Family Planning Pofisooofijvo§, l§(4), 157-162. Olson, L. (1980). Social and psychological correlates of pregnancy resolution among adolescent women: A review. Amarioan Joornal of Qrtnoooyoniaary, an, 432-445. Ortiz, C.G., & Nuttall, E.V. (1987). Adolescent pregnancy: Effects of family support, education, and religion on the decision to carry or terminate among Puerto Rican teenagers. aooloooonoa, 22, 897—917. Phipps-Yonas, S. (1980). Teenage pregnancy and motherhood: A review of the literature. Amorjcan Journal Of OELhooobelaLLy, an, 403-431. Resnick, M.D. (1984). Studying adolescent mothers’ decision making about adoption and parenting. Sooial Work, 23, 5-10. Richards, A.K., & Willis, I. (1983). unoor l§ and ogaonant. New York: Lothrop, Lee, and Shepard. Rosen, R.H. (1980). Adolescent pregnancy decision-making: Are parents important? aooloooonoo, 2;, 43-54. Rosen, R.H., Benson, T., & Stack, J. M. (1982). Help or hindrance: Parental impact on pregnant teenagers’ resolution decisions. Eamjly Bolationa, 21, 271-280. 142 Smith, E.W. (1975). The role of the grandmother in adolescent pregnancy and parenting. Ina Joornal Of §£h991.fl§§l&b. ii. 278-283- U.S. Department of Commerce (1980). 1980 Census Of Population: Classified Index of Industries and Occupations (1st ed.). Vincent, C.E. (1960). Unwed mothers and the adOption market: Psychological and familial factors. Maccjaga ang Eamily Livjng, 22, 112-118. Vincent, C.E. (1961). unmagojoo_moonona. Glencoe, IL: The Free Press. Wilt, R.L., & Michael, J.M. (1982). Mom, I’m orognant. New York: Stein and Day. Yelloly, M.A. (1965). Factors relating to an adoption decision by the mothers of illegitimate infants. Sociological_fiexieu. a. 5-14. Young, T., Berkman, B., & Rehr, H. (1978). Parental influence on pregnant adolescents. Sooja] Wonk, 29, 387-391. Zelnik, M., & Kantner, J.F. (1980). Sexual activity, contraceptive use and pregnancy among metropol'tan area teenagers, 1971-1979. Eamjly Planning Pacanootjves, 12(5). 230- Zelnik, M., Kantner, J.F., & Ford, K. (1981). So; ano r n n ' d s c . London: Sage Publications. nICHIan STATE UNIV. LIBRARIES llHllllllllIWWIWWWIllllllWlUlHlllWWI 31293008914420