BACKGROUND: This study was conducted to compare the utility of methods of assessing ambivalent attitudes about childbearing that require deductive reasoning by the subject to methods that do not. The goal was to predict the intent to use a noncoital method of contraception during adolescence. DESIGN: Participants (N=340) in a racially and ethnically diverse population (white 20%, black 25%, Hispanic 55%) completed two scales concerning attitudes toward childbearing-a traditional Likert scale and a scale with positive, negative, and "I go back and forth" response choices. The indication of ambivalence according to these two scales was determined by two methods-averaging item responses and counting the number of midscale responses. Logistic regression was used to study the relationship between ambivalence (by each method) and contraceptive plans. RESULT: Regardless of the scale's format (Likert or back-and-forth) and the scoring method used (averaging or counting), ambivalent adolescents were less apt to plan on using noncoital contraceptives than were nonambivalent adolescents. However, many of the adolescents who were classified as ambivalent by the averaging method chose no midscale responses (26.3% for the Likert scale and 40.5% for the back-and-forth scale), hence they were not classified as ambivalent by the counting method. These adolescents were younger and had lower average scores than adolescents who were classified as ambivalent by both scoring methods. Moreover, adolescents who were classified as ambivalent by both scoring methods were less likely to intend to use noncoital contraceptives than were nonambivalent adolescents, but adolescents who were classified ambivalent by only the averaging method were not. CONCLUSIONS: Childbearing ambivalence predicts contraceptive plans. However, congruent with theories of cognitive development, methods of assessing childbearing ambivalence that require deductive reasoning on the part of the adolescent tend to overclassify adolescents with discordant attitudes as being ambivalent. Avenues of further study are discussed. Copyright 2010 North American Society for Pediatric and Adolescent Gynecology. Published by Elsevier Inc. All rights reserved.
BACKGROUND: This study was conducted to compare the utility of methods of assessing ambivalent attitudes about childbearing that require deductive reasoning by the subject to methods that do not. The goal was to predict the intent to use a noncoital method of contraception during adolescence. DESIGN:Participants (N=340) in a racially and ethnically diverse population (white 20%, black 25%, Hispanic 55%) completed two scales concerning attitudes toward childbearing-a traditional Likert scale and a scale with positive, negative, and "I go back and forth" response choices. The indication of ambivalence according to these two scales was determined by two methods-averaging item responses and counting the number of midscale responses. Logistic regression was used to study the relationship between ambivalence (by each method) and contraceptive plans. RESULT: Regardless of the scale's format (Likert or back-and-forth) and the scoring method used (averaging or counting), ambivalent adolescents were less apt to plan on using noncoital contraceptives than were nonambivalent adolescents. However, many of the adolescents who were classified as ambivalent by the averaging method chose no midscale responses (26.3% for the Likert scale and 40.5% for the back-and-forth scale), hence they were not classified as ambivalent by the counting method. These adolescents were younger and had lower average scores than adolescents who were classified as ambivalent by both scoring methods. Moreover, adolescents who were classified as ambivalent by both scoring methods were less likely to intend to use noncoital contraceptives than were nonambivalent adolescents, but adolescents who were classified ambivalent by only the averaging method were not. CONCLUSIONS: Childbearing ambivalence predicts contraceptive plans. However, congruent with theories of cognitive development, methods of assessing childbearing ambivalence that require deductive reasoning on the part of the adolescent tend to overclassify adolescents with discordant attitudes as being ambivalent. Avenues of further study are discussed. Copyright 2010 North American Society for Pediatric and Adolescent Gynecology. Published by Elsevier Inc. All rights reserved.