Regional and urban analysts have developed graph-theoretic and multivariate approaches for decomposing the structure of flow tables. These developments have led to an important question: which method(s) should one use? In this study the author examines the ability of each of four approaches to generate industry groups (clusters) from a regional, survey-based input—output table. Specifically, the author deals with industry groups as generated by two graph-theoretic methods and by two multivariate techniques. The degree of similarity among the groups so generated is analyzed using measures of association. The composition of the groups depends on cutoff points. Nonetheless, the results suggest that one of the graph-theoretic methods and one of the multivariate techniques produce similar clusters. Further consideration indicates that these methods reveal more of the strongest interindustry linkages than their alternatives.
Get full access to this article
View all access options for this article.
References
1.
AnderbergM, 1973Cluster Analysis for Applications (Academic Press, New York)
2.
BeyersW B, 1972, “On the stability of regional interindustry models: The Washington data for 1963 and 1967”Journal of Regional Science12363–374
3.
BlinJ MCohenC, 1977, “Clustering and aggregation in input—output systems”Review of Economics and Statistics54376–380
4.
BourqueP JChambersE JChiuJ S YDenmanF LDowdleBGordonGThomasMTieboutCWeeksE, 1967, “The Washington economy: An input—output study”Graduate School of Business Administration, University of Washington, Seattle, WA
5.
CampbellJ, 1972, “Growth pole theory, digraph analysis, and interindustry relationships”Tijdschrift voor Economische en Sociale Geografie6379–87
6.
CampbellJ, 1975, “Application of graph theoretic analysis to interindustry relationships”Journal of Regional Science and Urban Economics591–106
7.
CzamanskiD ZCzamanskiS, 1977, “Industrial complexes: Their typology, structure and relation to economic development”Papers of the Regional Science Association3893–111
8.
CzamanskiS, 1971, “Some empirical evidence of the strengths of linkages between groups of related industries in urban—regional complexes”Papers, The Regional Science Association27137–150
9.
CzamanskiS, 1974, “Study of clustering of industries”, Institute of Public Affairs, Dalhousie University, Halifax, Nova Scotia
10.
CzamanskiSAblasL de Q, 1979, “Identification of industrial clusters and complexes: A comparison of methods and findings”Urban Studies1661–80
11.
EmersonJRinglebA, 1977, “A comparison of regional production structures”Papers, The Regional Science Association3985–98
12.
GiarrataniF, 1980, “The scientific basis for explanation in regional analysis”Papers, The Regional Science Association45185–196
13.
GiarrataniFMcNelisP, 1979, “Time series evidence on the structure of interindustry relationships”Review of Regional Studies945–53
14.
GoodmanL AKruskalW H, 1954, “Measures of association for cross classifications”Journal of the American Statistical Association49732–764
15.
HararyFNormanRCartwrightD, 1965Structural Models: An Introduction to the Theory of Directed Graphs (John Wiley, New York)
16.
HarriganF J, 1982, “The relationship between industrial and geographical linkages: A case study of the United Kingdom”Journal of Regional Science2219–31
17.
HooverE, 1975Regional Economics2nd edition (Alfred A Knopf, New York)
18.
HubertL J, 1973, “Min and max hierarchical clustering using asymmetric similarity measures”Psychometrika3863–72
19.
JensenR C, 1980, “The concept of accuracy in regional input—output models”International Regional Science Review5139–154
20.
JonesL P, 1976, “The measurement of Hirschmanian linkages”Quarterly Journal of Economics90323–333
21.
LathamW, 1977, “The measurement of spatial association”Review of Regional Studies720–30
22.
LeontiefW, 1965, “The structure of the US economy”Scientific American21225–35
23.
LodhE KLewisJ S, 1975, “Identification of industrial complexes from input—output tables of Canada and the USA: Some empirical tests”Empirical Economics153–80
24.
LoviscekA L, 1980, “Changing input—output coefficients: The case of West Virginia”Review of Regional Studies1014–28
25.
LoviscekA LHollidayR ERobinsonL AWolfordM A, 1979, “The West Virginia input—output study: Modeling a regional economy” (West Virginia University Press, Morgantown, WV)
26.
McGilvrayJ W, 1977, “Linkages, key sectors, and development strategy” in Structure, System and Economic Policy Ed. LeontiefWCambridge University Press, Cambridge) pp 49–56
27.
MasserIScheurwaterJ, 1980, “Functional regionalisation of spatial interaction data: An evaluation of some suggested strategies”Environment and Planning A121357–1382
28.
MostellerF, 1968, “Association and estimation in contingency tables”Journal of the American Statistical Association631–28
29.
RichardsonH W, 1975Regional Growth Theory (Macmillan, London)
30.
RoepkeHAdamsDWisemanR, 1974, “A new approach to the identification of industrial complexes using input—output data”Journal of Regional Sciences1415–29
31.
SlaterP B, 1977a, “Hierarchical internal migrations for Italy”Metron33182–187
32.
SlaterP B, 1977b, “The determination of groups of functionally integrated industries in the United States using a 1967 interindustry flow table”Empirical Economics21–9
33.
SlaterP B, 1978, “The network structure of the United States input—output table”Empirical Economics349–70
34.
SlaterP B, 1981, “Comparisons of aggregation procedures for interaction data: An illustration using a college student international flow table”Socio-economic Planning Sciences151–8
35.
StreitM E, 1969, “Spatial association and economic linkages between industries”Journal of Regional Science9177–193
