Fathers and Perinatal Loss Pt. 7: Method

Tracy Schaperow, Psy.D.

Tracy Schaperow

Licensed Clinical Psychologist

This article is part of a larger work.

Return to the table of contents | Go to the next section

Chapter III Method and Procedure


This study used a qualitative approach to examine five fathers after they had lost a baby through a neonatal death, stillbirth, or Sudden Infant Death Syndrome (SIDS). While this is a relatively small sample even for a qualitative study, it is important to keep in mind that the goal was not primarily of generalization. Rather, this researcher had decided in advance to use a limited number of participants in order to investigate and illustrate in maximum detail the themes that were discussed in the literature. The following inclusion and exclusion criteria were implemented to specify the focus and parameters of this study. Any individual who did not meet these criteria was given appropriate referrals (i.e., Psychotherapists in the community; support group).

Inclusion Criteria.

  1. Any married or single male who, within the last four years, had lost a baby less than one year old through Sudden Infant Death Syndrome (SIDS), ectopic pregnancy, miscarriage, neonatal death, or stillbirth was included in the study.

  2. Because this researcher is aware that the expression of grief depends upon one’s cultural traditions and practices, this study used participants from the same cultural and ethnic background (i.e., men of European decent born in the United States). Therefore, these findings do not generalize to men in other cultures.


Exclusion Criteria.

  1. Those individuals meeting the criteria for a Substance Abuse or Dependence diagnosis, not in sustained full remission (over one year sobriety), based on the American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders, Fourth edition, 1994 criteria, was excluded. All individuals were given a phone screen that assessed for substance abuse and dependence. If they met the criteria for either substance abuse or dependence, they were excluded from the study.

  2. Any individual showing evidence of a psychotic process (i.e., flight of ideas,

    ideas of reference, delusional thinking, hallucinations) in the initial screening or during the interviews, were excluded. Participants were asked about psychiatric hospitalizations and medications. If any of the potential participants exhibited any of the above signs or symptoms and are not stabilized on medication, they were excluded.
  3. Any individual at risk for suicide as determined by this evaluator by assessing the history of past suicide attempts and/or current suicidal ideation was excluded.

  4. Any individual who lost a child, who was at least one year old, was excluded from the study.


In the present study, an attempt was made to examine fathers’ conscious experiences and unconscious, defended, and indirectly expressed aspects of their experiences. Conscious experiences were obtained through semi-structured interviews and the Perinatal Grief Scale. This study investigated each participant’s unconscious experiences through the Thematic Apperception Test (T.A.T.), the human form of Children’s Apperception Test (C.A.T.-H), and a semi-structured interview. The following sections provide a more thorough explanation of each instrument.

Semi-structured Interview.

The interview had a semi-structured format, meaning that some of the questions were predetermined, and some of the questions were dependent on the direction of the interview. An interview guide (Appendix A) was created through the following process. First, the researcher decided on the main areas central to this study. These areas were determined by reviewing the literature on perinatal loss and through consulting with dissertation committee members. Second, this researcher wrote out questions. Third, the committee members read the questions and made suggestions for revisions. Then this researcher modified the questions. This process was repeated about four times. Last, a final interview guide was solidified.

In the interview, this researcher began by asking an open-ended question, and gradually moved toward more specific questions. The rationale for beginning with an open-ended question is that it allowed each participant to engage in conversation with the interviewer, and limited the use of yes or no responses (Maykut & Morehouse, 1994). By gradually working toward more specific questions, the researcher provided an opportunity for those men who had difficulty with open-ended questions to answer in a more comfortable manner. Furthermore, this procedure ensured that all of the pertinent questions about fathers and perinatal loss were addressed. The four main open-ended questions that were created for the interview are as follows:

  1. “Tell me about the loss,”
  2. “Tell me what you thought about having children before your partner was pregnant,”
  3. “Tell me what you thought having a baby would be like when you were growing up,”
  4. "Tell me about the pregnancy.”

Each of these questions had at least two or three levels of more specific questions that followed. For example, after asking the individual to Tell me what you thought of having children before your wife was pregnant, the researcher asked a slightly more specific question, such as, What were your thoughts and attitudes about having a baby, and then asked an even more specific question,

Some men feel excited, some feel unhappy, some feel afraid, and some men feel ambivalent about having a baby. What about with you? The interviewer was aware that some subjects require more prompting than others do; therefore, the interview guide was designed to adapt to the subjects’ needs for more or less questioning.

Essentially, the semi-structured interview aimed to gain information about the following topics: (a) the experience/circumstances of the perinatal loss, (b) the experience of the pregnancy for the father, (c) the desires and fantasies about having children, (d) the way life has changed/not changed following the loss, and (e) future plans.

Perinatal Grief Scale.

The short version of the PGS, (Appendix B), used in the current study, was created by Toedter and Lasker in 1988. The PGS is a 33-item Likert- type scale that is used specifically for pregnancy-related loss. It has items that vary from strongly agree (1) to strongly disagree (5). The original version of the PGS originally had 104 items. However, the final version had reduced the number of items to 84, thereby increasing the psychometric properties of the measure (Potvin, Lasker, & Toedter, 1989).

There were no clear criteria or closely related existing instruments with which to compare the PGS, so it was validated through a method created by Cronbach and Meehl in 1959 (Toedter Lasker, & Alhadeff, 1988). The Cronbach and Meehl method used constructs created in previous literature that were thought to be important in predicting perinatal grief and combined key constructs, resulting in a new instrument. As an 84-item scale, the PGS was shown to be reliable with a .97 alpha coefficient. Because the authors of the scale felt that it was still too long to be utilized by researchers and clinicians, they dropped items that were not highly correlated with most of the other items. This change resulted in the short version of the PGS, which has three subscales, each consisting of 11 items. Each subscale of the short version of the PGS has at least a .85 reliability alpha coefficient, suggesting that each subscale is internally consistent. Although the long version has been validated on men, the short version of the scale has not. (Potvin, Lasker, & Toedter, 1989). Moreover, only three men were willing to participate in the validation study of the long version of the PGS. Perhaps the men who did not participate in the study had reactions that were different from those of the men who were willing to participate. Thus, this instrument may not have accurately assessed the grief experiences among the fathers in this study. Due to these limitations of the PGS and the exploratory nature of this dissertation, the PGS was supplemented by other sources of information. That is, the semi-structured interview, the T.A.T., and the C.A.T.-H were also used.

The short version of the PGS contains three subscales entitled active grief, difficulty coping, and despair. These subscales were created to differentiate milder grief reactions from more debilitating and longer-lasting reactions from perinatal loss (Potvin, Lasker, & Toedter, 1989). Active grief, which is not considered to be severe, is also known as normal grief. This subscale includes questions regarding sadness, missing the infant, and crying for the baby. Difficulty coping suggests a hard time dealing with people and with activities. It is believed to indicate a more severe depression because the individual is withdrawing from others, and having trouble functioning with daily life. Despair indicates more serious and long-lasting detrimental effects of the loss. It is believed that the severity of an individual’s grief can be assessed based on his or her highest subscale score. Therefore, the three subscales can be seen to reflect a continuum of grief that ranges from active grief to despair, reflecting the particular view of grief of the PGS authors. Because Potvin, Lasker, and Toedter’s model of grief is not the only one in the literature, this dissertation used the PGS in an exploratory way to see if it is a useful dimension in capturing grief in men.

The T.A.T. and the C.A.T.- H.

The interview and the Perinatal Grief Scale were supplemented by selected cards of two projective instruments: the Thematic Apperception Test (T.A.T.) and the human form of the Children s Apperception Test (C.A.T.-H).

Bellak and Abrams (1997) described the evolution of the T.A.T. and the C.A.T.-H. According to these authors, Henry Murray created the T.A.T. in 1943. His instrument, still used today, consists of thirty black and white drawings that suggest various emotional themes (Bellak & Abrams, 1997). In 1948, Leopold Bellak and Sonya Sorel developed the C.A.T. for children because they thought that children might identify more with pictures of animals than with pictures of people. Because some children responded better to human figures than to animal figures, Bellak and Bellak modified the C.A.T. to human form (C.A.T.-H) in 1965.

The T.A.T. is based on the assumption that an individual s unconscious thoughts and feelings may be revealed in disguised ways through the stories told about these pictures. Numerous studies have used the T.A.T., however only a few of those studies have had topics that were similar to this dissertation. For instance, Ballou (1978) examined the significance of reconciliative themes during pregnancy in women. By using a projective device, such as the T.A.T., the study was able to look at the internal experiences of women during pregnancy. Similarly, Klatskin and Eron (1970) compared the projective content of pregnant women to their adjustment during the postpartum period. The T.A.T. enabled the researchers to examine the women s feminine identification as well as their acceptance of their pregnancy by looking for stories with positive female role themes. While most of this research has looked at women, the T.A.T. has been used in numerous studies on both men and women (Bellak & Abrams, 1997).

The C.A.T.-H, a descendent of the T.A.T., is an apperceptive test that was created by Bellak and Bellak in 1965 for children (Abrams & Bellak, 1997). In its original form, the C.A.T. included ten black and white plates depicting animals in diverse situations that elicited various emotional themes. The C.A.T.-H is the human modification of the C.A.T. It has pictures that suggest similar emotional themes to the C.A.T., however they have humans instead of animals in them. Analogous to the T.A.T., the C.A.T.-H is based on the assumption that an individual s unconscious thoughts and feelings may be revealed in disguised ways through the stories told to these pictures. The rationale for using this instrument as a supplement for the T.A.T. is that this researcher believes that the pictures of young children on the cards may evoke themes related to perinatal loss. Although this projective test was designed for children, there has been some speculation that it can also be useful for adults (Kitron & Benziman as cited in Bellak & Abrams, 1997). Because there is no research supporting the use of the C.A.T.- H for fathers after a perinatal loss, this researcher used this instrument as an experimental method.

Questions of reliability and validity depend heavily on the context and purpose for which they are being used. A variety of different coding scoring systems of satisfactory inter-rater reliability has been developed for the T.A.T. (Bellak & Abrams, 1997). Validity of various constructs measured by this test has also been examined in a variety of ways. Prior research on reliability and validity of these tests in not of central relevance in the present study, however, because this dissertation did not primarily aim to measure predetermined variables. Rather, this study used these instruments in a qualitative fashion to gain indirect information about participants’ feelings and reactions that they may be attempting to deny or defend against in the interview and the direct self-report measure. In this study, these instruments were used in the same fashion as any clinical evaluation.

The following section describes the specific parts of the instruments that were used in the present study. Three cards from both the T.A.T. and the C.A.T.-H were administered. The cards were chosen based on emotional themes related to this topic (i.e., children, loss, etc.). The primary investigator and the dissertation committee identified these emotional themes, which are largely based on the literature. Pictures 2, 3BM, and 7GF of the T.A.T. were used, and pictures 3, 5, and 9 of the C.A.T.-H were used. Picture 2 on the T.A.T. is a country scene in which a young woman holds books in the foreground. In the background of this picture, a man is working in the fields, and an older woman (who looks pregnant to some people) is looking at him. This card elicits themes concerning identity, family relations and possibly pregnancy (Bellak & Abrams, 1997). Picture 3BM has a boy huddled against a couch with his head leaning on his arm. There is a revolver next to him on the floor. This image typically elicits themes of anger, denial, and depression (Bellak & Abrams, 1997). Card 7GF has an older woman sitting beside a young girl on a sofa. The woman is reading or talking to the girl. The young girl is holding a doll on her lap and looking away. This picture evokes feelings about becoming a parent (Bellak & Abrams, 1997).

On the C.A.T.-H, card 3 has a picture of a man sitting in a chair with a pipe and a cane. On his right side, there is a little boy sitting on the floor looking up at him. This card aims to elicit themes about male identity. Card 5 has a picture of a dark room with two babies sitting in a crib. This card may elicit themes directly related to the loss of a baby. Card 9 has a picture of a dark room with a child in his bed/crib facing toward the door. This illustration pulls for themes of deserting a child or being deserted. The rationale for using these two projective devices is that they assess thematic content and interpersonal themes. Although systematic, far-reaching conclusions cannot be made from these projective instruments, these tests were used as another source of data and as an attempt to approximate more information than may be obtained from a clinical interview. Because these projective instruments have not previously been used to assess feelings surrounding perinatal loss in males, the use of them in this study was an experimental procedure. This study aimed to obtain information that future studies can test more systematically.

Visit the author at: PsychologyCT.com

comments powered by Disqus