Ageing in Sub-Saharan Africa: Tracing the elderly in population censuses - The example of Tanzania

Ageing in Sub-Saharan Africa:Tracing the elderly in population

censuses -The example of Tanzania

Doris SchmiedChair for Urban and Rural GeographyUniversity of Bayreuth, Germany

Elderly in Sub-Saharan Africa (SSA)

Rapidly growing numbers of elderly in AfricaRapidly changing life situations and roles of elderlyInformation on elderly very limited

Mainly empirical research in anthropology, sociology and health studiesSome practical development workScarce statistical information on elderly – except censuses

Census data in Sub-Saharan Africa

"Understandably, population censuses in statistically underdeveloped countries are the principal sources of information on a wide range of areas which are of vital importance to development planning."Yet"Utilization of the census information has been found to be minimal.“(Tanzania 1988 Population Census, The Analytical Report)

Census data in Tanzania

4 censuses after independence1967, 1978, 1988, 2002 all de facto censuses, all carried out in August

General problems with censuses in Sub-Saharan Africa

Logistic problemsincluding all households including all members of households

Enumeration stafftraininghonesty

Incorrect or misleading answers of interviewees because they find liaison with the enumerator unsatisfactory they are unaware of the significance/importance of their informationthey interpret terms used differently (multi-lingual situation)they cannot or do not wish to part with the correct information

Problems of tracing the elderly in the censuses

Problems with age in generalAge is not an unambiguous conceptBirthdays and years are not important in Africa

Problems with the definition of "the elderly"Most traditional African/Tanzanian societies are "gerontocratic" (although rapidly changing)Old age = senior position in society (Kisuahili mzee)Differences between sexes: • old men: loss of physical abilities, but apogee of economic/social power• old women: women after menopause, loss of child-bearing ability, status based on number of children (sons) or traditional knowledge, power over daughters-in-law Cultural diversity: e.g. age-set societies (horizontal bonding through rites of passage) 65+ : "past working life" - a European concept transferred to SSA

Difficulties with data on age in the censuses

Data on age may seem or is distorted because:

• Age data is collected directly and indirectly

• Interviewed people may be unaware of their own age• Household heads may not know the age of the members of their household• Age stated is influenced by intentions of the interviewed • Age stated reflects digital preferences: tendency to rounding/heaping• Age stated is influenced by enumeration procedure: cards used to facilitate the identification of age predispose answers

• Major events influence age distribution

Digital preference, Tanzania

Figure 1A: Diagramatic Representation of Age Structure for the 2002 Census: Tanzania.

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

Age in Single Years

Perc

enta

ge

Both Sexes Male Female

Digital preference, Tunduru District

Digital preference: Dodoma Rural, Dodoma Region

Influence of major crisis?1921, 1947, 1954 famine1991 Cholera outbreak (57 people die in Ndogorowe Village alone)1974, 1986, 1998 serious food shortages

AgeBoth Sexes Male Female

TOTAL 438866 207706 231160

65 2920 1136 155666 789 326 46367 1091 554 53768 1930 772 115869 741 302 439

65 - 69 7243 3090 415370 3755 1631 212471 474 219 25572 1284 586 69873 505 218 28774 588 289 299

70 - 74 6606 2943 366375 1416 689 72776 729 342 38777 377 213 16478 1018 490 52879 416 226 190

75 - 79 3956 1960 199680+ 5299 2306 2993

Age in the Tanzanian census

Can data on age be used at all?

What about data on "elderly"?

Age ratios, Tanzania 1967, 1978 and 1988 censuses

Sex ratios, Tanzania 1967, 1978 and 1988 censuses

Can data on age be used at all? What about data on "elderly"?

Data on age can be used becauseage distortion in old age is hardly more pronounced than at younger ageage distortions have followed a similar pattern in all censuses: hardly any changes over timebroad age groups level out distortions to a great extent

Important to keep data weaknesses in mind!data on elderly women are more distorted than on elderly mensmall size of old age groups means that data tends to exaggerate tendencies

The 2002 Population and Housing Census

Information largely available on the netIncludes detailed data on ageNew: district data - District ProfilesNew: expanded questionnaire (socio-economic data)

Testing the 2002 Tanzania Population and Housing Survey

1. Proportion of elderly2. Sex ratios of elderly3. Ageing in the city and the countryside4. Marital status of elderly5. Disability among elderly

Proportion of elderly (65+ years) by district, Tanzania Mainland, 2002

Sex ratios of all elderly (65+ years) by district, Tanzania Mainland, 2002

Sex ratios of elderly (65 - 69 years) by district, Tanzania Mainland, 2002

Sex ratios of elderly (70 - 74 years) by district, Tanzania Mainland, 2002

Sex ratios of elderly (75 - 79 years) by district, Tanzania Mainland, 2002

Sex ratios of elderly (80+ years) by district, Tanzania Mainland, 2002

65-69 70-74

75-79 80+

0

20000

40000

60000

80000

100000

120000

0 to 4 5 to 9 10 to 14 15 to 24 25 to 34 35 to 44 45 to 54 55 to 64 65+

males females

Ageing in the city - Dar es Salaam, 2002

Ageing in the countryside - Tunduru District, 2002

Marital status of elderly

0102030405060708090

100

nevermarried

married livingtogether

separated divorced widowed

males females

0102030405060708090

100

nevermarried

married livingtogether

separated divorced widowed

males females

Ngorongoro District Tunduru District

0

2

4

6

8

10

12

14

0-4

5-9

10-1

4

15-1

9

20-2

4

25-2

9

30-3

4

35-3

9

40-4

4

45-4

9

50-5

4

55-5

9

60-6

4

65-6

9

70-7

4

75-7

9

80+

males females

Disability among elderly

0

2

4

6

8

10

12

14

0-4

5-9

10-1

4

15-1

9

20-2

4

25-2

9

30-3

4

35-3

9

40-4

4

45-4

9

50-5

4

55-5

9

60-6

4

65-6

9

70-7

4

75-7

9

80+

males females

Ngorongoro District Tunduru District

Results

Recent census data from Sub-Saharan Africa can and should be used to gain more information on elderly (e.g. IPUMS census data on Kenya and South Africa)Researchers have to be aware of the considerable shortcomings

BUTCensus information on elderly is more valuable than guestimates of international organizationsCensus information allows regional differentiation Weaknesses of data do not prevent the formulation of working hypothesesCensus information on elderly forms an important basis for empirical research