| dc.description.abstract | This study was generally focused on evaluating the contribution and extent to which the impact of Family Planning Programmes have been effected to reduce fertility in Kenya by employing some of the existing impact assessment models and to identify some of the conceptual and measurement problems inherent in each model that account for the variation of results.
More specifically, the objectives include estimating the births averted at National level and Provincial level by Program contraception.. Non-program contraception and type of method using the Standardization model, the Prevalence model and the Multivariate regression model.
Three sources of data namely Census 1989, DHS-I (1989) and DHS-II (1993) were utilized according to data requirement specified by each model. The following are some of the findings obtained
The results of the evaluation by the standardization model indicate that, although, marital status increased in Central, Eastern and Nairobi more births were averted in Central, Eastern and Rift Valley provinces (see summary results at the end of chapter five) On the other hand, although, marital status decreased between 1989 and 1993 in North Eastern Province, marital fertility, increased and hence births were added rather than averted. Both marital status and marital fertility indicate an increase between 1989 and 1993 in Nairobi and hence additional births, although too small, are observed to have been added. However. this anomaly may have arisen from the projected data for 1993
In terms of figures, according to the standardization approach, the total explained change brought about by the four components of CBR at National level is -798 births (a decline) per 1000
population. When this decline is applied to the 1993 projected population, about 207.447 births were
averted. Of this figure births averted by the changes in marital fertility (which in turn is attributable to the impact of family planning programmes) is about 157,052 births At provincial level. Central Province recorded the highest number of births averted caused marital fertility decline (about
291 percent of the total change explained in the province) followed by Eastern Province (about 117 percent of the total change explained in the province) followed by Rift Valley (about 74 percent) Generally, the study found that the standardization approach with some adjustments to incorporate births out of marriage, could be a good instrument for evaluating the impact of family planning
programs on fertility as it uses the real population data.
Results of the evaluation by the aggregate version of the prevalence model indicate that the highest number of births averted occurred in Central Province (about 37 births per 1000 women) followed by Nairobi (about 21 births per 1000 women) and Coast Province (about 20 births per 1000 women) while the lowest number of births are averted in Western Province; while the lowest number of births were averted in Nyanza Province (about II births per 1000 women) and Western Province (about 7 births per 1000 women). However, this result obtained from 1993 DHS could not be compared with that of DHS, 1989 since the aggregate version was not applied then.
The evaluation by the age-specific version of the prevalence model indicates that program contraception averted more births in 1993 than 1989 - an increase of about 38%; while non-program contraception averted less births in 1993 than in 1989 - a decrease of about 35% which shows that while program contraception had improved in performance, non-program contraception had decreased at National level. Generally, the study found that at National level age groups 25-29. 30-34 and 35-39 recorded prevalence rate over 35 percent while age group 15-19 is the least contraception age group.
Higher births were averted in age groups 30-34. 25-29 and 35-39 in order of descent At provincial
level, higher number of births were averted in Central (about 37 births per 1000 women) followed by Easter Province (about 29 births per 1000 women) while the least was Coast (about 17 births per 1000 women) for the same age groups
The evaluation by the method-specific version of the prevalence model revealed that among program methods, Pill, IUD, injectable were used more in Nairobi, Central and Eastern than the other provinces On the other hand, abstinence is practiced more in Nairobi, Eastern, and Central in order of descent More births are averted by pill in 1993 than in 1989 in all regions. Condom is poorly prevalent in all regions ranging from 0.6 percent in Nyanza to 2.33 percent in Central Province and births averted by the method is negligible compared to other methods.
The results of the analysis by the Multiple linear regression model on 1989 and 1993 DHS data indicate that the family planning component (use of any method of FP) explained the variability of the dependent variable (CEB) better than respondent's work status and electricity, radio and TV facilities: while socio-economic indicators, literacy level, toilet facility, husband's education and husband's occupation explained the dependent variable better than use of any method of family planning. Although the regression models obtained by this study do not exhaustively explain the dependent variable due to left-out variables, the family planning program component was not the most important predictor even when compared with the few socio-economic and demographic variables included in the models. The study found out that it is inappropriate to deduce that in both 1989 and 1993, family planning played the leading role in explaining the variability in the dependent variable, CEB. in other words, use of any method of family planning did not play the leading role in reducing fertility. Indeed, socio-economic indicators have likely played a better role, although the result is susceptible due to the
tact that all possible variables that are related to the dependent variable were not included m the multiple linear regression analysis.
Limitation of the Standardization mode! are described in detail at the end of chapter five. Accordingly, 15.5 percent of births in 1989 and about 37 percent of births in 1993 are considered negligible since the model does not incorporate births-out-of-marriage. This study suggests the introduction of an adjustment factor in order to incorporate births with-in and out-of marriage. Other suggestions for future study and policy making include: assessment of the impact of family planning on both fertility and mortality, that more deeper investigation, particularly operational research be carried out for the improvement of the prevalence of Condom use as it has dual advantage - for spacing or
averting births and protecting from STDs and HIV/AIDS infection. | en |
