By Abate Mammo
December 1992
Accurate and timely economic and demographic data are essential for fair and efficient allocation of resources as well as for acceptable electoral representation. Sensible public debates and compromises among various interest groups should no longer be hampered by a false sense of demographic reality.
Are the Oromo the majority ethnic group in Ethiopia? Is Ethiopia predominantly Muslim? Is Wello region predominantly Oromo? Is the urban population in Eritrea and Gonder more literate than the urban population in Bale, Wellega and Illubabor? According to the 1984 census, the answer to all these question is “no”. The picture that emerges from the 1984 population census reflects both conventional wisdom and surprise: Ethiopia is a largely Christian, a culturally diverse and regionally balanced “nation of ethnic minorities.”
Statistical Data
Until very recently little was known about the population of Ethiopia. The First and Second Round National Demographic Surveys provided valuable estimates of the size distribution of the population of Ethiopia by some basic characteristics such as age and gender. In 1981, a survey on the sedentary population of rural Ethiopia provided a still better demographic information for the surveyed areas. Areas not surveyed for security problems included Eritrea, Tigrai and some other awrajas.
The 1984 national census revealed a population count of 42,616,878, some 2,859,262 more than what was expected based on earlier surveys. The census had provided Ethiopia the first and most complete account of its people. For the first time also, the distribution of the census population by important factors such as age, gender, religion ethnicity, geographic area and many other characteristics of the population are now available.
Ethiopia currently stands the third most populated country in Africa after Nigeria and Egypt and ranks 20th in the world. Its population is characterized by high mortality, high fertility, high unemployment (and high under-employment) and is one of the fast growing nations (in population size: growth rate of 2.9 percent per year) in the world. In 1992, for example, the population of Ethiopia (including Eritrea’s 3,387,500) is estimated at about 55 million and is expected to reach 71 million by the year 2000.
Some Specifics
The 1984 census provided us with data on 92 ethnic groups in Ethiopia many of which were small in size (some even as small as the Kewama {283 people} and Sheta ({88 people}). The largest ethnic group, the Oromo, accounted for 29 percent (12,387,674) of the total population and was closely followed by the Amhara which accounted for 28 percent (12,055,250). The distant third and forth largest ethnic groups, respectively, were the Tigrawai and Gurage.
The Oromo are proportionately under-represented in the urban areas as are the Amhara and many other smaller groups. For example, while the Oromo account for 29 percent of the total population, they make up for only 7 percent of the urban population. The Amhara account for 28 percent of the total population but only 18 percent of the urban population. On the other hand the third and fourth largest ethnic groups, the Tigrawai and the Gurage, were over-represented in the urban areas. The most urbanized ethnic groups in Ethiopia consisted of the Adere (82 percent of 29,518) and the Dorze (62 percent of 43,964). We clearly see that no single ethnic group in Ethiopia is a majority and that Ethiopia is indeed a nation of many minorities which, depending on the region they live in, may constitute a majority.
The most populated region was Shewa (including Addis Abeba) accounting for 19 percent of the total population followed by Harerge (10 percent). Sidamo and Wello each accounted for 9 percent of the population each while Gojjam accounted for 8 percent and Gonder for 7 percent of the population. The regions that had the smallest share of the population were Illubabor and Bale (2 percent each). The observed ethnic mix in each region exhibits the diversity of our nation and the relative harmony that must have existed for centuries.
Ethiopia is one of the least urbanized countries in the world with only 11 percent of its people living in urban centers. With the exception of a few cities, males in many urban areas are outnumbered by females. The higher gender ratio in Asseb, Wonji, Shewa and Dilla may be explained by the male dominated migration to these cities relative to others. Ethnic groups also tend to have different sex patterns of migration to urban centers. For example, in Addis Abeba there were 34 percent more Gurage males than Gurage females when all other major ethnic groups (e.g., the Amhara, the Oromo and the Tigrawai) were outnumbered by females. This differential sex-specific ethnic migration was, in part, responsible for the observed population distribution of Addis Abeba.
Over 1.4 million residents were reported in Addis Abeba in 1984 and accounted for 29 percent of the total urban population. The population was 50 percent Amhara, 17 percent Gurage, 17 percent Oromo, 8 percent Tigrawai and the rest consisted of different ethnic groups. Addis Abeba inhabitants were 86 percent Christian with Muslims accounting for only 11 percent of its residents.
The national religion distribution was consistent with expectation. We find, for example, that 61 percent of Ethiopians were Christian, 33 percent Muslim and that 6 percent of our population practice indigenous religions. Consistent with expectation again, Gojjam, Gonder, Wellega and Shewa regions were predominantly Christian while Harerge and Bale were predominantly Muslim. The surprise regions, however, were Arsi, Illubabor, Kefa and Wello where the differences between the proportion Christian and the proportion Muslim were much smaller than one would have expected based on conventional wisdom. Gamo Gofa and Sidamo had a substantial proportion of indigenous religion followers. Indigenous religion was also practiced by a sizable proportion of populations in Kefa and Illubabor residents.
Since the literacy campaign started early in the 1980s its “effectiveness” has become the subject of discussion. Before the census date only 8 percent were reported literate in Ethiopia. In 1984 about 19 percent reported that they were attending schools making the total percent literate as high as 27 percent. The literacy report assumed that those attending schools at the time of the census would have been able to read and write in the language of instruction. This, however, is not necessarily true and literacy status might have been exaggerated. Literacy status in Ethiopia varied by region, gender and rural or urban residence. In rural areas, for example, males were about 2.4 times more likely to be literate whereas in urban areas the gap between the sexes was smaller (1.3 times). As expected, rural areas were the least literate with Gamo Gofa (12 percent) and Kefa (17 percent) falling at the bottom and Arsi and Bale at the top, with 30 percent and 35 percent, respectively.
The increase in the percentage of literate was the result of the literacy campaign. An assessment of the contribution of the literacy campaign vis a` vis the direct and indirect costs incurred on the campaign will have to consider the growth of the school age population, the expansion of schools as well as many other relevant factors, such as duration of the program. Literacy status of the population at present is expected to be much higher than 27 percent. Such rapid increase in literacy status, supplemented
by the observed decline in the observed population-to-medical doctor ratio will translate into a better standard of living in the long run through, for example, higher income and better health.
Many seem to be pre-occupied with questions that relate to ethnicity and religion more than common problems which all our people share including chronic malnutrition, unprecedented poverty, ill health, lack of education and very high unemployment. A careful study of our society will reveal that no single social group has the edge over these and other social indicators. For example, Gojjam which is populated by 87 percent Amhara and 95 percent Christian had one of the highest infant mortality rates (IMR) in 1981 (182 infant deaths for every 1,000 live birth). Wellega, on the other hand, is 89 percent Oromo and 85 percent Christian but had only half the IMR (95 per 1,000) of Gojjam. Gamo Gofa which has a negligible proportion of the largest ethnic groups (the Amhara and the Oromo) and splits its population into 49 percent Christian and 48 percent indigenous religion followers recorded an IMR of 162. This example clearly shows that homogeneity in religion and/or ethnicity (or their lack of) does not explain regional IMR differences not will it inform us of the country’s unacceptably high IMR rate of about 150.
Sometimes differences in social indicators do not always happen in the same direction. An example of this may be obtained from a child mortality study conducted by the author on the settled population of rural Ethiopia. Controlling for literacy status, religion and other background factors, it was found that in the North (Shewa, Gojjam, Gonder and Wello), Amhara children died more often than expected compared to Oromo children. In the rest of the regions (excluding Eritrea and Tigrai) Oromo children died more often than Amhara children. Such differences are brought about by complex social, economic and cultural factors that are not easily reduced to one or two indicators. The public debate should, therefore, be more diverse in its coverage of issues so that we will educate ourselves more about the marvels of our rich cultural heritage. An inaccurate understanding of our demographic realities will only hinder any social reform efforts. Ethnicity and religion should only be discussed in their capacity as proxies for other socio-economic indicators. We may start by expecting writers to substantiate claims. The census and other surveys may serve as starting points in this regard. It is time that we refer to factual data that are within acceptable margins of error when disseminating information on our population. Only then will one tell fact from fiction.
_________________
Abate Mammo, Ph.D., is a demographer by training. He currently works for the State of New Jersey Department of Health as Research Scientist. Statistical tables and references have been omitted.