Reproductive change in Ghana : evidence from two national surveys
Date
1994
Authors
Amenuvegbe, Benjamin E
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This analysis examines various aspects of human reproductive change based on the
Ghana Fertility Survey (GFS) and the Ghana Demographic and Health Survey (GDHS).
It compares the data on transition to parenthood, birth spacing and durations of
breastfeeding, postpartum amenorrhoea and abstinence and the factors influencing
them. The study also dealt with the use of modem contraception as well as the
relationship between the survival status of a child and the time to the next birth. The
consistency in the measurement of selected variables in both surveys was also
examined.
There was considerable agreement in the measurement of birth intervals and the effects
of factors affecting them. Consistency has been demonstrated in the measurement of the
second, fifth and sixth birth intervals and also in the effect of selected independent
variables examined in the two surveys. Age at first marriage, first birth and the first
birth interval do not appear to be comparable in the two surveys; this is also the case
with the third and the fourth birth intervals. This finding is not surprising because of
problems of measurement of marriage data in Africa which result from the fact that it is
often a process rather than an event at a point in time.
The finding of inconsistency of measurement of age at marriage and the first birth
interval implies that these data may not be pooled from the GFS and the GDHS to obtain
a larger and possibly more stable series; also it is necessary to exercise caution in
interpreting observed differences when such data are compared for the GFS and GDHS.
On the other hand the consistency found shows reliability of the data though at aggregate
level, and thus considerably increases our confidence in the comparative results.
The study found general stability in the pace and quantum of fertility. The overall
distribution of the second birth interval is only marginally different for the two surveys;
first, third and fourth interval distributions are not significantly different between surveys. The distribution of the second to the fourth intervals are fairly symmetrical though the
younger cohorts appear to show slightly longer delays for these births.
Age at first marriage and age at first birth also appear to be generally stable within and
between surveys though small increases are observed for later periods and younger
cohorts. The only significant differential for age at first birth is education: women with
11 or more years had higher ages at first birth in both surveys.
Median durations of breastfeeding and postpartum sexual abstinence either are stable or
have increased slightly between the GFS and GDHS. A positive relationship is found
between birth interval length and breastfeeding duration and also between fertility on the
one hand and amenorrhoea and abstinence on the other indicating the importance of
these postpartum factors for fertility. Median durations of postpartum amenorrhoea are
found to generally exceed those of postpartum sexual abstinence. This suggests that
amenorrhoea is more important for birth spacing than sexual abstinence among the
respondents.
Since failure time data from large-scale surveys are usually heavily tied, the grouped
proportional hazard model is applied to determine important factors influencing age at
first birth, birth intervals as well as breastfeeding and postpartum amenorrhoea and
abstinence. Ethnicity, place of residence and education are found to be significantly
related to breastfeeding and postpartum sexual abstinence in both surveys.
Finally, the analysis showed a positive relationship between breastfeeding and birth
interval length which indicates the contraceptive effect of breastfeeding, and implies that
the long periods of breastfeeding in Ghana are advantageous not only for the sake of
infants' health but also for birth spacing. Since effective contraceptive use is low, and the
women tend to bear children throughout the reproductive life-span, long breastfeeding
durations are important for fertility reduction. This analysis concludes that there is only a weak evidence for reproductive change. Few
and generally weak differentials were observed within and across surveys. Also, lack of
significant association between postpartum variables and age and parity indicates stability
of fertility decline. Except for the weak evidence mainly associated with recent periods
and also younger cohorts, the pace and quantity of fertility appear to be stable.
Description
Keywords
Citation
Collections
Source
Type
Thesis (PhD)
Book Title
Entity type
Access Statement
License Rights
Restricted until
Downloads
File
Description