In Exercise 9.1 we saw data on infant mortality and a number of other variables. There you predicted infant mortality from income. There is reason to believe that infants of young mothers are at increased risk, and there is considerable evidence that infant mortality can be reduced by the use of contraception. Does the multiple regression using all three of those predictor variables bear out these hypotheses?

Exercise 9.1

In Sub-Saharan Africa, more than half of mothers lose at least one child before the child’s first birthday. Below are data on 36 countries in the region, giving country, infant mortality, per capita income (in U.S. dollars), percentage of births to mothers under 20, percentage of births to mothers over 40, percentage of births less than two years apart, percentage of married women using contraception, and percentage of women with unmet family planning need.
(http://www.guttmacher.org/pubs/ib_2-02.html) Available at the book’s website as SubSaharanInfMort.
dat. (Notice the effect of outliers.)

Transcribed Image Text:

Need % mom % mom <2 yrs Using Country InfMort Income <20 > 40 apart contraception family planning Benin Rep 104 933 16 Burkina Faso 109 965 17 Cameroon 80 1,573 21 Central African Rep 102 1,166 22 Chad Rep 110 850 21 Cte d'Ivoire 91 1,654 21 Eritrea 76 880 15 55 455367 5 17 17 25 26 24 3573- 26 26 20 3. 16 missing 6 16 4 28 7 26 4 28 Ethiopia 113 628 14 Gabon 61 6,024 22 Ghana 61 1,881 15 Guinea 107 1,934 22 Kenya 71 1,022 18 Madagascar 99 799 21 645535 20 6 23 4 22 12 28 13 13 23 17 4 24 23 5 31 Malawi 113 586 21 Mali 134 753 21 64 6 17 4 26 Mozambique 147 861 24 Namibia 62 5,468 15 Niger 136 753 23 Nigeria 71 853 17 Rwanda 90 885 Senegal 69 1,419 14 Tanzania 108 501 19 646 9 6755775 19 22 25 5 27 21 18 2025525a38 24 10 26 26 30 26 7 26 9 13 17 17 27332 17 18 36 35