Research Repository UCD

No generally accepted definition of social geography exists. The variety of literature which has appeared under the title of social geography is astounding; even within particular schools there are wide disparities of approach and definition. With some notable exceptions, for example, in Sweden and Holland, social geography can be considered a field created and cultivated by a number of individual scholars rather than an academic tradition built up within particular schools. Furthermore, for many people the term “social geography” itself is in disfavor because its past association with various forms of determinism that postulated a causal connection between society and the geographical environment.

Perhaps, therefore, the best way to examine social geography is to establish a general theoretical outline of the field and, on this basis, to review the existing literature.  Naturally, many of the works relevant to what is here called social geography will have been written as contributions to some other discipline

Buttimer, Anne

Title Social Geography
Authors(s) Buttimer, Anne
Publication date 1968
Publication information Buttimer, Anne. “Social Geography.” The Macmillan Company & The Free Press, 1968.
Publisher The Macmillan Company & The Free Press
Item record/more 
information
http://hdl.handle.net/10197/10722
Downloaded 2024-05-27 16:06:18
The UCD community has made this article openly available. Please share how this access
benefits you. Your story matters! (@ucd_oa)
© Some rights reserved. For more information















Social Geography

An examination of the effect of sex and age structure on psychiatric hospitalization rate and its regional variations in Ireland has been carried out. Statistical testing, by regression, of the influence of selected socioeconomic variables on hospitalization rate produced some interesting results. Of most interest was the finding that elderly populations tended to increase hospitalization rate at all age groups.

It is hoped to carry this work further using more refined data

Walsh, Dermot

Walsh, Brendan M.

English

Some influences on the intercounty variation in Irish psychiatric hospitalization rates

Title Another look at the Concept of "Overpopulation"
Authors(s) Walsh, Brendan M.
Publication date 1968-10
Publication information Walsh, Brendan M. “Another Look at the Concept of ‘Overpopulation’” 17, no. 1 (October, 1968).
Publisher The University of Chicago Press
Item record/more 
information
http://hdl.handle.net/10197/1494
Downloaded 2024-05-27 13:39:05
The UCD community has made this article openly available. Please share how this access
benefits you. Your story matters! (@ucd_oa)
© Some rights reserved. For more information
Another Look at the Concept of "Overpopulation" 
Brendan M. Walsh 
Tufts University 
Kamerschen's concept of "overpopulation" seems to have survived the 
initial discussion in this journal and elsewhere.1 This apparent vitality 
seems to me to be attributable rather to lack of cogency in the criticisms 
that have been voiced than to the soundness of the original idea. Be- 
fore too much effort is expended on the construction of Kamerschen's 
index or too much weight placed on the conclusions based on its use, 
it might be valuable to try to discuss some of the idea's fundamental 
weaknesses. 
Briefly, Kamerschen suggests that "overpopulation" may be simply 
and fairly reliably measured by constructing an index defined as the Total 
Dependency Ratio (TDR), where TDR = (100) (population aged 65 
and over + population aged under 20)/(population between 20 and 65). 
Apart from the ease with which the TDR may be constructed, the follow- 
ing virtues are claimed for this index: first, "overpopulation" as defined 
by this index is a better measure of "population 'pressure' "2 than any 
other single statistic available; secondly, in some ways the TDR may 
represent an improvement over the income-per-person measure of economic 
welfare, primarily because the TDR has certain "dynamic" properties 
supposedly absent from the income-per-person concept3 (from the terms 
of Kamerschen's subsequent argument, these dynamic properties appear to 
be the superior predictions about an economy's future growth rates that 
may be made on the basis of the TDR). The TDR should provide a rule-of- 
thumb test for the existence of what we all presumably agree to call popu- 
lation "pressure," and it should also be useful as a supplement to the 
national-income-per-person concept, especially for making predictions 
about the economy's future. Some arbitrary critical level for the TDR has 
to be set, and countries may then be classified as "overpopulated" or "not 
overpopulated" on the basis of the values of their TDR. Kamerschen 
tentatively sets the critical value at 100, above which lies overpopulation. 
No implication is made about the extent of overpopulation. 
Although not relying on theoretical justifications for the use of the 
TDR, Kamerschen does suggest some reasons for its validity.4 First, he 
95 
Economic Development and Cultural Change 
suggests that those aged over 65 and under 20 are relatively immobile and 
less able to adjust to economic incentives than the population aged between 
20 and 65. While this conclusion is probably justified for those aged over 
65, it is certainly not justified for those aged, say, between 14 and 19: 
empirical studies of migration have found that that mobility is generally 
high for those in the latter age group.5 Secondly, it is asserted that 
Intuitively it seems reasonable to feel that as the number of economically 
unproductive workers increases relative to the productive group, i.e., the 
dependency ratio increases, a country would have more of a population 
problem.6 
The implication is that the TDR measures the ratio of "unproductive 
workers" to "productive workers" in a population. Of course, the TDR 
was not defined in terms of the age distribution of the labor force, nor in 
terms of the ratio of "productive workers" to "unproductive workers," 
but simply in terms of the age distribution of the total population. To 
assume that the age distribution of the total population reflects the distri- 
bution of the population between producers and nonproducers fully evades 
the crucial problem of defining and measuring the productive labor force 
in an underdeveloped country, a problem which has preoccupied many 
economists in recent years. Damaging though this fault may be, it is at 
least defensible as a crude approximation which would have the virtue of 
making the measurement of "overpopulation" much easier, if less mean- 
ingful. To accept, however, without comment Bogue's boundaries of 
demarcation between active and dependent population is inexcusably 
arbitrary and casual: Bogue's index was for use as a demographic device for 
the United States, and it properly reflects the present U.S. conventions 
with regard to entry into and retirement from the labor force.7 To apply 
this index to the underdeveloped countries and to hope that economic 
meaning can be extracted from it is clearly to sidestep all the issues that 
attend the definition and measurement of overpopulation. 
Of course, Kamerschen does not place too much emphasis on the 
theoretical justifications for the use of the TDR; instead, he emphasizes 
that his index has considerable support from the available empirical data, 
although he admits that it may be possible to find exceptions to his claims 
about it.8 Faced with this sort of methodological vacuum, it may be difficult 
to disprove anything to Kamerschen's satisfaction, since his thesis has been 
hedged to such an extent that it may not admit of proof or disproof. A 
further look at the data is in order, however. 
In the various cases he discusses in which the TDR test suggests that 
"overpopulation" does not exist, while the income per person test suggests 
that the nation is underdeveloped, Kamerschen is pleased to report that 
in almost all cases the countries in question performed well in terms of 
growth of real income per person in the subsequent period, and this ap- 
96 
Brendan M. Walsh 
parently supports the claims made for the superior "dynamic" qualities 
of the TDR. While not wishing to get involved in an example-counter- 
example contest, least of all when it entails the comparison of demo- 
graphic and economic statistics for the underdeveloped countries, it may 
be useful to point out that since the TDR is a purely demographic con- 
struct, it is possible for it to rise above the critical level as a result of forces 
that are not clearly related to any usual concept of "overpopulation" nor 
to the future course of national income. A country with European-level 
death and birth rates, a slowly growing, stationary, or even falling total 
population could assume the demographic profile necessary to yield a 
TDR > 100, if heavy and persistent emigration occurred in the 20-65 
age group of its population. There is nothing in this set of circumstances to 
suggest overpopulation, unless one is willing to accept emigration as a 
sign of this problem (Kamerschen explicitly rejects this), nor does it imply 
anything about the future course of the country's income. All that is 
necessary is that a substantial gap exist between this country's income and 
the income of some other country to which immigration is not restricted. 
The sufficient condition is that the country's population should have 
reacted to this income gap through heavy and persistent emigration that 
was selective of those in the 20-65 age group. 
In point of fact, the only country in which these circumstances have 
combined in recent history is Ireland, although migration of sufficient 
intensity may well have given many regions within the larger, richer nations 
the necessary population profile. (In fact, Bogue records a TDR of 99.9 
for the rural farm population of the United States in 1950.9) Between 
1956 and 1961, the population of the Republic of Ireland fell at an annual 
rate of 0.56 percent: the annual crude birth rate was 21.2, the crude death 
rate was 11.9, and the net emigration rate was 14.8 per 1,000 population.10 
Almost 80 percent of this emigration was from the population aged 20-65. 
In 1961 the TDR for Ireland was 102.4 (or 100.8 on Kamerschen's earlier 
definition), thus just passing the "overpopulation" test, despite the falling 
population, despite being the country with the lowest population density 
in Europe, and despite the $740 level of income per person in 1964. From 
the "dynamic" viewpoint the TDR fares no better in this case: Ireland's 
real income per person grew at an annual average rate of 4 per cent in the 
four years after 1961. The unusual Irish population structure is, of course, 
a serious problem and raises very interesting economic and social prob- 
lems.l1 Emigration from Ireland may, in fact, be related to the very high 
fertility rates that are prevalent there, as well as to the low level of income 
in Ireland relative to the United Kingdom. But these are separate issues, 
involving a more detailed discussion of the concept of "overpopulation" 
than this debate has provided to date. The example of Ireland is adduced 
here merely to underline the fact that it is theoretically possible and his- 
torically verified that a country may pass Kamerschen's test and still not 
be overpopulated in any of the ways in which this term is generally used. 
97 
Economic Development and Cultural Change 
The Irish case also serves to provide a counterexample to the "dynamic" 
claims of the TDR. 
The upshot of this discussion is inevitably negative. The only con- 
clusions that appear to emerge relate to research strategy: instead of looking 
for easy ratio tests to enable us to define "overpopulation," the most 
profitable course would seem to be to construct theoretical models of the 
interrelations between population growth and income growth and to test 
these models by established econometric techniques. 
1 Cf. David R. Kamerschen, "On an Operational Index of'Overpopulation,'" 
Economic Development and Cultural Change, Vol. 12, No. 1 (January 1965), pp. 
169-87; and Gordon A. Marker, "Comments on an Operational Index of 
'Overpopulation,"' Economic Development and Cultural Change, Vol. 15, No. 2 
(April 1967), pp. 336-38, with a "Reply" by Kamerschen, pp. 338-41, as well 
as the references cited in the last two items. 
2 Kamerschen, "On an Operational Index ...," op. cit., p. 171. 
3 Ibid. 
4 Ibid., p. 174. 
5 For a bibliography of migration studies, cf. Gunnar Olsson, Distance and 
Human Interaction, A Review and Bibliography (Philadelphia: Regional Science 
Research Institute, 1965). 
6 Kamerschen, "On an Operational Index . . .," op. cit., p. 174. 
7 According to footnote 8 of Kamerschen's "Reply" (op. cit.), he is not really 
sure just how he wishes to define the TDR: "Presumably this group [those aged 
exactly 65] should be included in the numerator." (In his calculations this group 
was completely omitted from both numerator and denominator!) In my defini- 
tion of the TDR I have taken this footnote into account, but in the calculation 
of the TDR I give the figure that results from both definitions. 
8 Kamerschen, "On an Operational Index . ..," op. cit., p. 186. 
9 Cf. Donald J. Bogue, The Population of the United States (Glencoe, Ill.: 
Free Press, 1959), p. 102. 
10 Cf. Ireland, Central Statistics Office, Census of the Population of Ireland 
1961, Vol. 2, Table 10; and for national income data, Statistical Abstract of 
Ireland, 1965, Table 235. 
11 For examples, cf. P. Kaim-Caudle, Social Policy in the Irish Republic 
(London: Routledge and Kegan Paul, 1967). 
Volume 17 Number I 
? by the University of Chicago 
All rights reserved 


Another look at the Concept of "Overpopulation"

Title Hospitalized psychiatric morbidity in Ireland : a suggested approach
Authors(s) Walsh, Dermot, Walsh, Brendan M.
Publication date 1967-06
Publication information Walsh, Dermot, and Brendan M. Walsh. “Hospitalized Psychiatric Morbidity in Ireland : A 
Suggested Approach” 113, no. 499 (June, 1967).
Publisher Royal College of Psychiatrists
Item record/more 
information
http://hdl.handle.net/10197/1490
Downloaded 2024-05-27 13:42:13
The UCD community has made this article openly available. Please share how this access
benefits you. Your story matters! (@ucd_oa)
© Some rights reserved. For more information
Independent
Variable%
Population
aged 65+Net
(out-)
Migration
RateDependent
Variable
HR 1959 .. ..
FAR .. .. ..,
4.5@*
261324@247HR
sehizophrenia . .
HRdefective. . ..4.02* 4@95*3
.37*3.@*HRmanicdepressive
HRsenile .. ..2.11 I@o93.Io@o76HR
neurosis . . . .i . 77@70
Brit. 3. P.@ychiat.(1967), 553, 675â€”676 ABSTRACT
T@ata I
Values of t-Ratiofor Simple Regression Coqificients
(sign ÃœZdiCaIeSnatureof relation)
Ireland has the highest rate of hospitalized mental
illness on record. This is borne out both by the
World Health Organization figures for treated
mental illness (W.H.O., ig6i), by international
comparisons of point prevalence rates (Arentsen and
Stromgren, 1959) and by consideration of first ad
mission rates (Hammer and Leacock, 1961). The
Irish national figures for each of these measures were
respectively7 .12, ioâ€¢8iand i 45 per 1,000popula
tion in 1959. The rates exhibit a distinct geographical
pattern within Ireland: they are highest on the
coastal areas furthest from Dublin, while in the
Dublin area they arc as low as those of most western
European countries. All three measures have i4sen
over time, and the intercounty variation has also
increased since i@oo. In 1959 the coefficients of
variation for both first admission and hospitalization
rates were equal to 25 . I per cent.
Because information from an individual patient
reporting system established in ig63 is not yet
available, our data on hospital admissions and
populations are crude and summary. In particular,
we are handicapped by the absence of age- and sex
specific rates by county. However, the data that are
available serve as a basis for some preliminary
hypothesis-testing.
We believe that socio-economic factors have an
important role to play in the explanation of the Irish
hospitalization and first admission rates and their
intercounty variation. It is impressionistically clear
that the regional pattern of these rates and of certain
socio-economic variables is similar. From i9oo to
â€˜¿959these rates increased throughout the country,
but this increase was most rapid in counties with
high rates of net (out-) migration and high ratios of
the aged to total population (partly a consequence of
population loss through migration). These counties
are mainly in the west, south-west and north-west.
In order to explore this relationship further we
have employed a regression model. We believe that the
data lend themselves to testing in this manner.
Table I shows the values of the t-ratio for the simple
regression coefficients of hospitalization rate by
S For a discussion of the statistical properties of re
gression models the mathematically oriented reader is
referredtoGeary (i96@).
S denotes Pt < .@ if true regression coefficient =o.
diagnosis on two socio-economic variables. The
results substantiate the impression of the relation
between these variables and HR for schizophrenia
and mental defect and for first admission rates, but
not for HR for senility and neurosis. A further test
shows no significant relationship between the
percentage of population aged over 65 and the
percentage of an area's hospital patients aged over
65. It seems likely, therefore, that the relationships
ofTable I would hold also for age-standardized rates.
Table II presents some multiple regression results.
We have introduced the level of expenditure on
mental health per person in an area as a measure of
the availability of hospital facilities in that area. We
suggest the following commentary on the results.
The HR increases with the percentage of persons
aged over 65 and with expenditure per person. The
net influence ofthe age structure on HR schizophrenia
remains significant after the introduction of the
expenditure variables. A high net migration rate is
associated with a high HR for schizophrenia, mental
defect and manic depressive psychosis and a low rate
for neurosis. This may support the hypothesis that
patients in the first three diagnostic categories
migrate less frequently than the population at large,
while those in the fourth migrate more frequently.
However, the results shown for HR schizophrenia
regressed on â€œ¿%Population over 65â€• and â€œ¿Fax
penditure per Personâ€•show the difficulty of separa
ting the age-structure influence from the migration
675
Hospitalized Psychiatric Morbidity in Ireland:
A Suggested Approach
By DERMOT WALSH and BRENDAN WALSH
Independent Variables%
Population
aged over 65Net
migration
rateExpenditurep rhea R2Dependent
VariableHR
1959 .. ..
FAR â€˜¿959.. ..2@85 1262@73 i@86@â€¢67.35HR
schizophrenia . .2@ 75*I@OI.53HR
schizophrenia . .2@23Iâ€¢¿33@48HR
defective . . . .2@ 63(3.97.49HR
manic depressive . .2 @8g@â€”o@ 58.3'HRsenil
....â€”0292.28*22HRneurotic
.. ..3.29*252.47
676 HOSPITALIZEDPSYCHIATRICMORBIDITY IN IRELAND: A SUGGESTEDAPPROACH
TABLE II
Values of t-ratiofor Regression Coefficients,Selected Multiple Regressions
S denotes pt< 05 if true regression coefficient=o.
influence with the available data : including both
variables in the one regression usually led to the age
structure variable remaining significant and the
migration variable becoming insignificant. The need
for data standardized by age, sex, marital status and
urban/rural residence is evident. Finally, it appears
that hospitalization for senility and neurosis increases
with the increased availability of facilities, as meas
ured by expenditure. The results for first admission
rates lend support to the overall interrelation
between the socio-economic variables selected and
this measure of the incidence of mental illness.
In summary it may be said that the first admission
and hospitalization rates are highest in areas that
have experienced high out-migration and have a
high percentage of their population aged over 65.
The variable â€œ¿%aged over 65â€•appears to affect
hospitalization in some way other than through the
higher rate of hospitalization among the elderly. The
association between migration rates and hospitaliza
tion by diagnosis suggests some interesting relation
ships, and may be related to other findings in this field.
The conclusions we have drawn from our data arc
tentative and refer to the Irish experience; cross
cultural validity is not claimed. The ability ofthe data
to support any hypotheses regarding different
diagnostic groups is limited by the sharpness with
which these groups are defined and diagnosed.
Although we believe that first admission rates are
meaningful in relation to severe mental illness, it is
realized that hospitalized mental illness is not coter
minous with mental illness in the community. We
have, however, corrected at least partially for
regional variations in the difference between illness
and hospitalization through the use of a measure of
the availability of mental health care.
We claim that the present approach holds promise
in the study of the remarkably high rates and
variation of rates of hospitalized mental illness in
Ireland, for which no explanation has been offered so
far (Walsh, 1962). We feel that socio-economic
variables exert a considerable influence upon mental
illness in Ireland. The present results, however, are
based upon inadequate data and are therefore of use
only in establishing very broad interrelations. As
more and better data become available it is hoped to
pursue these topics in greater detail.
REFERENcES
ARENTSEN, K., and STRÃ”MGREN, E.@ â€œ¿Patients in
Danish psychiatric hospitals.â€•Acta Jutlandica, 131, I.
MediCal Series g.
Ga*ay, R. C. (1963). â€œ¿Somer marks on relations
between stochastic variables : a discussion document.â€•
Rev.mt.Stat.Inst.,35,2,163â€”181.
HAMMER, M., and La@cocx, E. (i@6i). â€œ¿Sourcematerial
on the epidemiology of mental illness.â€•In: Field
Studies in the Mental Disorders, 418-476. Ed. Zubin, J.
New York. Grime and Stratton.
WALSH, D. (1962). â€œ¿Cultural influences in psychiatric
illnesses in the Irish.â€•3. Irish Med. Arm., 297, 62-68.
World Health Organization. (i96i). Ep@nie1Ogied and
Vital Statistics Report. Geneva, i@Ã§,5.
Dermot Walsh, M.B., D.P.M., Acting Physician Superintendent, St. Lomans Hospital, Palmerstown, Co.
Dublin
Brendan Walsh, M.A., Ph.D., Assistant Profissor of Economics, Tufts University, MeJford, Massachusetts
02155, U.S.A.
(Received â€˜¿7May, 1966)


University College Dublin

Not a member yet

Social Geography

Some influences on the intercounty variation in Irish psychiatric hospitalization rates

Another look at the Concept of "Overpopulation"

Hospitalized psychiatric morbidity in Ireland : a suggested approach

12,080

15,234