Chapter 3: Research Proposal
The lack of long-term follow-up of women post-abortion along with no control groups, lack of standardized measures, and the high drop-out rate of women participating in post-abortion research are just some of the methodological weaknesses mentioned above. Denial and repression may also prevent the woman from dealing with the effects of the abortion until years afterwards, during which time she may feel that the procedure had no adverse effects (Boyland, 1992; Major et al, 2000; Franz & Reardon, 1992; Congleton & Calhoun, 1993; Lewis, 1997; Doherty, 1995; Butler, 1996).
With all of the methodological weaknesses in mind, the current study proposal has been constructed. The purpose of this study focuses around post-abortion syndrome and its prevalence in women who have had abortions in the United States. The Millon Clinical Multiaxial Inventory III (MCMI-III) quantitatively measures characteristics of women who have not had abortions (GP3) and those who have had an abortion 10-15 years previously (GP2) and those who had an abortion within the past year (GP1). Open-ended interview questions assess the qualitative dimensions involving the thoughts and feelings of women who have experienced abortion. This study seeks to assess the types of adverse effects women experience and to what extent these effects continue to be prevalent 10-15 years after the abortion. The following hypothesis will be examined:
- Post-abortion distress (PAD) symptoms (measured by the MCMI-III scales 2B, D and A) will be significantly higher in GP1 than in GP2 or GP3.
- PAS symptoms (measured by MCMI-III scales B, T, R, and CC) will be significantly higher for GP2 than for GP1 or GP3.
- Post-abortion psychosis (PAP) symptoms (measured by the MCMI-III scales 1, S, SS, and PP) will be significantly higher for women in GP1 and GP2 who have experienced second or third trimester abortions.
- Socio-economic and cultural factors (measured by ethnographic evaluation of directed interviews (see Appendix C)) are more often associated with negative symptoms in GP1 and GP2 than in GP3.
Geographical locations selected for the research study proposed include the midwest, south, east, and west regions of the United States. Specific states from each of these areas will be selected according to the level of interest met by organizations within their state that meet the criteria for the present proposal.
The primary investigator will contact mental health organizations in each region. Networking will be done through the director of the facility to form the research groups of women.
The interviewee receives information explaining the purpose of this research project, and is asked to participate by their health care provider. All the ethical guidelines outlined by the American Psychological Association (APA) will be integrated into the process of selecting and informing participants regarding the proposed project.
Upon arrival for the interview, the female participant is asked to read and sign a consent form, which reviews the experimental design and purpose and also serves as a confidentiality statement between the interviewer and the participant. Then, the interviewer fills out a personal information form that will provide demographic data for statistical analysis (see Appendix F). The interviewer then administers the MCMI-III and proceeds into the open-ended interview questions upon completion of the MCMI-III (a five to ten minute break may be taken before continuing the interview). The open-ended quality of the interview questions can lead to some information being shared during each interview that is unique to that particular woman’s story. The format has been created with flexibility that allows the interviewer to make adjustments based on each woman’s response.
Building rapport and an environment that fosters the sharing of the participant’s story is necessary to facilitate openness and honesty with the interviewer. The participant is debriefed afterwards and, if necessary, referred to a counselor or given other references for follow-up on issues discussed during the interview. There is also a post-interview evaluation form to allow the woman to reflect on the interview process, the interviewer, and her level of honesty throughout the course of the interview (see Appendix E). The women in GP1 will be invited to participate in a longitudinal study that will follow them for ten years after the initial interview process.
The study will involve 144 women: forty-eight that have experienced an abortion up to one year ago, forty-eight women who have experienced an abortion ten to fifteen years ago, and forty-eight women who will serve as a comparison group that have not had an abortion. There will be no age restrictions on the women who participate in the study, although an effort will be made to have equal representation of age groups in each sample group. Women selected to contribute will be from different regions of the United States (midwest, west, east, and south) and will be contacted through the mental hospital or health clinic in their area. GP1 participants will be invited to join in the longitudinal investigation that will continue for a ten year period.
Theodore Millon created the MCMI-III as a diagnostic tool to be used in conjunction with the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV). The test is designed to be used with adults eighteen and older and takes about twenty-five minutes to complete. There are 175 true-false items yielding scores on 24 total content scales for each participant (see Appendix D). This study focuses upon differences in the Post-Traumatic Stress Disorder, Dysthymia, Alcohol Dependence, Drug Dependence, and Anxiety scales across the three participant groups (Reardon, 2000).
According to a review in the Mental Measurements Yearbook, the MCMI-III has the highest internal consistency of any test in the industry (varying from .66 to .90). The test-retest reliability coefficients are also high with a testing period of five to fourteen days. Some weaknesses of the test include high codependence between the scales and thirty-nine of the 175 items are keyed false leaving the MCMI-III vulnerable to someone who answers yes to a significantly high number of questions (Impara & Plake, 1998).
The interview that follows the MCMI-III is made up of thirteen open-ended interview questions asking about the abortion procedure, post-abortion attitudes, sources assessed before making the abortion decision, and other prevalent information (see Appendix C). These questions seek to attain qualitative data about the women’s experience. MCMI-III data is compared to interview data in order to assess different facets of the woman’s life that have the potential to be affected by the abortion. An explanation regarding the selection of the interview questions is found in Appendix C.
The researcher recognizes that the projected direction of these interview questions may not be followed and women being interviewed could cover
information that the interview process was not aiming to measure through a particular question. The open-ended quality of these questions does not invalidate the value of the information that will be attained.1
The proposed research project seeks to compare the results of the MCMI-III among the female participants who have not had abortions (control group) and among those who have had abortions within the past year or 10-15 years (experimental groups) and look for significant differences in the scales.
The first research question will focus on results from scales 2B, D, and A on the MCMI-III for women one year after the abortion experience compared to those 10-15 years after the abortion and the comparison group.
The second research question suggests significant differences in the MCMI-III scales B, T, R, and CC for women who experienced an abortion 10-15 years ago compared to women who had an abortion one year ago and the comparison group.
The research question proposes significance in the results of the 1, S, SS, and PP scales of the MCMI-III for both experimental groups of post-abortion women compared with the control group. For the first three hypotheses comparisons, frequencies, and t-scores will be used to assess significant differences at the .05 level.
The fourth research question will examine the interview questions and the significant differences in responses based on socio-economic and cultural factors for women post-abortion 10-15 years compared to the women one year post-abortion. An ethnographic assessment, the NUD*IST software, will be used to assess possible significance in these characteristics. Demographic information will also be collected to control for differences among the control and experimental groups which may be responsible for any differences in the hypothesized directions. Significant differences in age, race, economic status, marital status, number of other children, home state and religious affiliation will be assessed using independent samples t-tests at the .05 level.
The data will be assessed with a multivariate analysis in order to find nonlinear functions across the groups of women being studied.
1 Qualitative methods follow an open systems approach and are usually not presented in numerical form. The objective is to gather a variety of potentially useful information, while recognizing that the information gathered depends on the approach employed and varying circumstances, available resources, and the researchers running the investigation (Dunn, 2001).