Reasons why many women chose to have abortions

Subject: HARD DATA: Why U.S. women have abortions
Summary: Also, why they have *late* abortions

Source: "Why Do Women Have Abortions"
by Aida Torres and Jacqueline Darroch Forrest
Family Planning Perpectives, 20 (4) Jul/Aug 1988, pp 169-176
(The bimonthly research journal of The Alan Guttmacher Institute)

[Posted without my editorial comment, except for this one:

    "Sex selection" abortions almost never occur in the U.S.  Nowhere in
    the article do the authors mention sex selection.  None of the
    several professionals who helped design the survey apparently
    thought sex selection was worth considering.  Neither did the women
    who were surveyed.  At the beginning of the questionnaire -- before
    answering any "canned" questions -- the women were asked to write
    out their reasons for having an abortion.  From their un-prompted
    answers, the authors created another new category for reporting the
    results, "not ready for the responsibility."  Yet, the authors did
    not create a "sex selection" category.  In other words, none (or
    almost none) of the surveyed women mentioned this reason, either.

 I have included all of the five tables in the paper (while exercising
 great care to transcribe the numbers correctly!), along with a few
 explanatory excerpts.  My interjections are all enclosed within
 square brackets. -- Dean Benjamin]

SUMMARY [in its entirety]

Most respondents to a survey of abortion patients in 1987 said that more
than one factor had contributed to their decision to have an abortion;
the mean number of reasons was nearly four.  Three-quarters said that
having a baby would interfere with work, school, or other
responsibilities, about two-thirds said they could not afford to have a
child and half said they did not want to be a single parent or had
relationship problems.  A multivariate analysis showed young teenagers
to be 32% more likely than women over 18 or over to say they were not
mature enough to raise a child and 19% more likely to say their parents
wanted them to have an abortion.  Unmarried women were 17% more likely
than currently married women to choose abortion to prevent others from
knowing they had had sex or became pregnant.

Of women who had an abortion at 16 or more weeks' gestation, 71%
attributed their delay to not having realized they were pregnant or not
having known soon enough the actual gestation of their pregnancy.
Almost half were delayed because of trouble in arranging the abortion,
usually because they needed time to raise money.  One-third did not have
an abortion earlier because they were afraid to tell their partner or
their parents that they were pregnant.  A multivariate analysis revealed
that respondents under age 18 were 39% more likely than older women to
have delayed because they were afraid to tell their parents or partner.


A total of 819 abortion facilities in the United States performed at
least 400 abortions each in 1985; they represented 31% of all abortion
providers, but accounted for 90% (1.4 million) of all abortions.
Only such facilities were eligible for the study [...].

Some 42 facilities were originally invited to participate in the study;
these include six at which a relatively large number of late abortions
(those at 16 or more weeks' gestation) were performed.

Over a five-month period beginning in November 1987, patients at each
general abortion facility were surveyed during a 1-3-week period (the
length depended on the facility's caseload).  Providers of later
abortions were asked to administer the survey to 60 consecutive
patients.  [Omitted: Reasons for eliminating some data; overall response
rates.]  In all, 1900 women responded with useful information, of whom
420 had been pregnant for 16 or more weeks.  Because such women had
been oversampled, their reasons for having an abortion were weighted
to reflect the proportion of U.S. abortion patients who obtain
midtrimester abortions.

[Omitted: Comparison of survey data with known distributions of age,
race, ethnicity, Medicaid status, marital status, and region.]
Distributions of respondents by age and marital status were quite
similar to those of all abortion patients, but 17% of respondents were
covered by Medicaid, compared with 24% nationally.  There was little
difference among regions in the most important reasons given for
obtaining an abortion, but there were differences by race and by
Medicaid status.  In an attempt to obtain a more appropriate
description of the reasons U.S. women have abortions, responses were
weighted by these two factors, as well as by the gestational
distribution of abortions.


Information was collected by means of a self-administered questionnaire
distributed by clinic staff to patients.  The survey instrument was
reviewed by a number of professionals with some experience in the
provision of abortions and was pretested with some 150 patients at seven
facilities.  It was made available in both English and Spanish.  The
questionnaire covered both the women's reasons for choosing to have an
abortion and (for those at least 16 weeks pregnant) their reasons for
having delayed obtaining an abortion.

Both to ensure that no reasons were missed and to find out which were
most important, respondents were also asked to write, in their own
words, why they were having an abortion; if they had more than one
reason, they were requested to note the most important first.  This
question preceded the precoded questions, so as not to influence
respondents' replies.

Almost all of the responses to the open-ended question fell into one
of the 13 precoded categories.  However, some women said in their
replies to that question that they were choosing to have an abortion now
because they were "not ready for the responsibility of having a child."
As is duscussed below, these women did not clearly fall into any one of
the 13 categories, so their response was left as a separate category.

TABLE 1.  Percentage of abortion patients reporting that a specific
reason contributed to their decision to have an abortion, by age, and
percentage saying that each reason was the most important.

                                       |Total|            A G E          |Most
                                       | N = | <18 18-19 20-24 25-29 >=30|Impt
R E A S O N                            |1900 |(275)(309) (645) (337)(319)|1773
                                       |     |                           |
Woman is concerned about how having a  |  76%|  92%  82%   75%   72%  69%| 16%
baby could change her life             |     |                           |
                                       |     |                           |
Woman can't afford baby now            |  68 |  73   73    70    64   58 | 21
                                       |     |                           |
Woman has problems with relationship or|  51 |  37   46    56    55   50 | 12
wants to avoid single parenthood       |     |                           |
                                       |     |                           |
Woman is unready for responsibility    |  31 |  33   40    36    25   18 | 21
                                       |     |                           |
Woman doesn't want others to know she  |  31 |  42   41    35    21   22 |  1
has had sex or is pregnant             |     |                           |
                                       |     |                           |
Woman is not mature enough, or is      |  30 |  81   57    28     7    4 | 11
too young to have a child              |     |                           |
                                       |     |                           |
Woman has all the children she wanted, |  26 |   8   12    23    31   51 |  8
or has all grown-up children           |     |                           |
                                       |     |                           |
Husband or partner wants woman to have |  23 |  23   29    25    18   20 |  1
abortion                               |     |                           |
                                       |     |                           |
Fetus has possible health problem      |  13 |   9   13    12    14   17 |  3
                                       |     |                           |
Woman has health problem               |   7 |   3    4     7     8   15 |  3
                                       |     |                           |
Woman's parents want her to have abortn|   7 |  28   12     4     3    2 |<0.5
                                       |     |                           |
Woman was victim of rape or incest     |   1 |   1    1     1     1  <0.5|  1
                                       |     |                           |
Other                                  |   6 |   2    5     8     5    8 |  3
Notes: 1. N's [parenthesized counts in each age category] are unweighted
          [by race, Medicaid status, and gestational distribution].
       2. The N's upon which the age breakdowns are based do not add to 1900
          because age was not available for some women.

[Here is a summary of Table 1, as it appeared in the popular press.
 This compact version is more convenient for quoting in discussions:]

Source: 1987 survey of 1900 abortion patients
        by the Alan Guttmacher Institute.
        Published in the LA Times, Mon 17-Apr-89, Part I page 16
        Each woman was allowed to cite more than one reason.

   76   Concern about how having a baby could change her life
   68   Can't afford a baby now
   51   Problems with relationship or desire to avoid single parenthood
   31   Not ready for responsibility
   31   Doesn't want others to know she has had sex or is pregnant
   30   Not mature enough, or is too young to have a child
   26   Has all the children she wanted, or has all grown-up children
   23   Husband or partner wants her to have abortion
   13   Fetus has possible health problem
    7   Woman has health problem
    7   Woman's parents want her to have abortion
    1   Woman was victim of rape or incest
    6   Other

TABLE 2.  Percentage of respondents offering various additional details
for each of the three leading reasons women gave for having an abortion.

     Unready for how having a baby could change her life (N=1339)

67%  A baby would interfere with job, employment or career
49   A baby would interfere with school attendance
28   Children or other people depend on her for care

     Can't afford baby now (N=856)

41%  Woman is student or is planning to study
22   Woman is unmarried
19   Woman is unemployed
14   Woman has low-paying job
 9   Woman can't leave job
 7   Woman is on welfare
 6   Woman's husband or partner is unemployed
 5   Woman can't afford basic needs
 4   Woman receives no support from her husband or partner

     Problems with relationship or with single parenthood (N=790)

49%  Woman doesn't want to marry partner
32   Couple may break up soon
29   Partner doesn't want to or can't marry
25   Woman is not in a relationship
 6   Woman's husband or partner mistreats her or children
 5   Woman is unready to commit herself to a relationship

TABLE 3.  Multivariate regression coefficients
[This table, together with its interpretation, are listed after Table 5.]

TABLE 4.  Percentage of women who reported that various reasons
contributed to their having a late abortion and who cited specific
reasons as accounting for the longest delay

 All   delay
(399)  (311)  Reason

 71%    31%   Woman didn't recognize she was pregnant or misjudged gestation
 48     27    Woman found it hard to make arrangements for abortion
 33     14    Woman was afraid to tell her partner or parents
 24      9    Woman took time to decide to have an abortion
  8      4    Woman waited for her realtionship to change
  8      2    Someone pressured woman not to have abortion
  6      1    Something changed after woman became pregnant
  6    <0.05  Woman didn't know timing is important
  5      2    Woman didn't know she could get an abortion
  2      1    A fetal problem was diagnosed late in pregnancy
 11      9    Other

TABLE 5.  Among women who provided additional information relating to
three specific reasons for having abortions at 16 or more weeks'
gestation, percentage who gave various detailed reasons for delay.

     Woman failed to recognize pregancy or misjudged gestation (N=277)

50%  She didn't feel physical changes
50   She hoped she was not pregnant
33   She had irregular periods
32   She thought she had her period
20   Her MD underestimated gestation
20   She was practicing contraception
 9   Her pregnancy test was negative
 7   She didn't know where or how to get a pregnancy test

     Woman found it hard to make arrangements for an abortion (N=185)

60%  She needed time to raise money
32   She tried to get an abortion from a different clinic or MD
26   She had to arrange transportation because there was no nearby provider
20   She didn't know where to get an abortion
16   She couldn't get an earlier appointment
11   She took time to notify her parents or get their consent
 9   She needed child care or a Medicaid card
 0   She needed time to obtain court permission

     Woman took time to decide to have an abortion (N=74)

78%  She found having an abortion to be a difficult decision
19   She had religious or moral reasons for waiting
11   She talked with her parents/husband/partner

TABLE 3.  Multivariate regression coefficients (unstandardized) showing
association between selected independent variables and reasons for
choosing to have an abortion.

                   |                          R E A S O N
                   |             Don't want
                   |Unready for  sex activity               Have      Parents
Independent        |change       or pregnancy               enough    want
Variable           |in life      known          Not mature  children  abortion
Age < 18           |   --         -0.091          0.315       --       0.186
Age >= 30          |   --           --           -0.126      0.103      --
Unmarried          |  0.110        0.166           --         --       0.039
Hispanic           |   --           --             --       -0.079      --
Student            |  0.178        0.139          0.096       --        --
Employed           |  0.069        0.052         -0.055       --      -0.033
No children        |  0.094        0.121          0.269     -0.432     0.032
>=16 weeks         |   --           --             --         --       0.032
Catholic           |   --          0.077           --         --      -0.028
No religious affil |  0.060         --             --         --        --
Covered by Medicaid|   --         -0.095           --         --        --
No prev abortions  |   --          0.083          0.047       --        --
R**2 [R squared]   |  0.105        0.114          0.293      0.295     0.096
Note: Only coefficients significant at p < 0.05 and equations for which
the R**2 was 10% or greater are shown in this table.  Measures of
poverty status (<100% and 100-149% of poverty level) and of race (black)
were included in each equation, but are not shown here because they were
not significant.  In 1987, the federally designated poverty level for
a nonfarm family of four was $11,200.

A multivariate analysis was conducted so that numerous variables could
be taken into account simultaneously.  Table 3 shows the results of
regression analyses incorporating a wide range of personal
characteristics that might affect the likelihood that a specific
reason contributed to a woman's decision to have an abortion.  All
variables used in these analyses are expressed as dummy variables --
coded as "1" if a respondent has the characteristic and as "0" if she
does not.  Thus, the unstandardized coefficients shown in Table 3 can
be interpreted as showing the likelihood that a woman with a given
characteristic (net of all other characteristics, or independent
variables, in the analysis) will report a specific reason for having
an abortion.  Again, standard tests of significance have been used as
a rough guide, and only the regression coefficients with less than a
5% probability of occurring by chance are shown.

Regression analyses were conducted separately for each of the 13
reasons shown in Table 1, using as dependent variables both whether a
woman cited a specific reason at all and whether the reason was the
most important one.  Results of the two analyses were similar, so
Table 3 shows outcomes based on whether a specific reason was cited at
all.  The factors included in the analysis explained anywhere from
0.4% to 29.5% of the variance in the likelihood that each reason would
be cited, an indication that other factors not included in the
analyses had a substantial influence in determining whether a woman
cited a specific reason.  The results shown in Table 3 exclude
equations in which less than 10% of the variance was explained.  All
regressions were run using unweighted data, since the factors used to
calculate weights (race, Medicaid status and gestation) were included
as independent variables in the equations.

Table 3 shows that women under 18 were 32% more likely than those 18
or older to have decided to obtain an abortion because they weren't
mature enough to raise a child, and they were 19% more likely to have
elected to have an abortion because their parents wanted them to do
so.  Women 30 and over were 10% more likely than those younger than 30
to have made such a decision because they did not want to have more
children and 13% less likely to have done so because they thought they
were not mature enough.

A surprising result is that women under 18 were less likely than older
women to say that concern about others knowing that they were having
sex or that they werre pregnant was a factor in their decision.  Table
1 shows this factor to have been cited by nearly one-third of all
women and by two-fifths of those under 20.  However, given the
negative coefficient seen in the multivariate analysis, factors other
than young age must have contributed to the higher proportion observed
among those under 20.  This also may reflect the fact that young
adolescents concerned about others knowing are underrepresented among
abortion patients because they are less likely to confide in an adult
and to receive help in arranging access to services.

Unmarried women were 17% more likely than currently married women to
have chosen to obtain an abortion to prevent others from knowing they
were having sex or had become pregnant, and were 11% more likely to
have done so because having a child would interfere with other plans
or responsibilities.

Surprisingly, race and poverty status were not significantly related
to any of the reasons.  Although women with an income of less than
200% of poverty were significantly more likely to say they were having
an abortion because they could not afford to raise the child, the
regression explained only 5% of the variance in the range of those
giving that reason.

It was expected that students would be more likely than nonstudents to
be concerned about the changes a baby would bring or to feel that they
lacked the necessary maturity, but students were also 14% more likely
not to want others to find out about their sexual activity or
pregnancy.  Roman Catholic respondents were 8% more likely than those
with other religious beliefs to be having an abortion because they did
not want others to know, and were 3% less likely to say that their
parents wanted them to have an abortion.