Changeability of social-economic
factors in the development of
population transformations of
Silesian towns
Bulletin of Geography. Socio-Economic Series nr 5, 51-64
2006
B U L L E T IN OF G E O G R A P H Y (SOCIO-ECONOMIC SERIES) NO. 5/2006
A n n a R u n g e
Un iv e r s it y of Sil e sia in So sn o w iec
CHANGEABILITY OF SOCIAL-ECONOMIC FACTORS IN THE DEVELOPMENT OF POPULATION TRANSFORMATIONS
OF SILESIAN TOWNS
ABSTRACT. The paper concerns the assemblage of towns of historical Silesia studied in three patters: 1) as a set of 183 urban centres, 2) in regional system (Lower Silesia, Opole Silesia, Upper Silesia), 3) in size-class pattern (large towns: >100 tho usand inhabitants, medium: 20-100 thousand, small: <20 thousand). An attempt was made to identify correlations between the population number and features, which signi ficantly influence population changes. These factors include the assemblages of demo graphic, infrastructural, economic and financial features. The analysis was performed for four time cross-sections, which represent various political, economic and social situ ation of Poland (years: 1977, 1985, 1994, 2001). The employed method of step regres sion made it possible to identify three clear trends of correlations between population and the selected social-economic factors:
— in the scale of the whole territory of Silesia these correlations are rather small, but increasing in time (both before and after political transformation);
— in regional pattern, there is a clear increase of correlations from the west to the east; — in the pattern of size-classes of towns - the strongest correlations occur in large
towns, weaker in medium towns and the weakest in small towns.
KEY WORDS: Silesia, population transformations, step regression, factors of so cial-economic transformations.
Urban settlement system o f the historical area o f Silesia shows the occur rence o f several types o f settlement system, which developed within the attach ment to different countries and within administration borders o f different pro vinces. They include Upper Silesian conurbation, agglomeration system o f Opole
Silesia and Lower Silesian agglomeration. The latter shows the supremacy of Wrocław and weaker developed middle hierarchical level o f towns. These three settlement systems differ from each other in terms o f the size structure o f towns and also trends of their demographic development (Runge A., 2004a).
The investigations concern the assemblage o f towns o f Silesia understood as a historical-geographical region (Drobek, 1999), which are analysed in the pattem o f individual dimension groups of towns (large, medium, small) on the background o f regional system (Lower Silesia, Opole Silesia, Upper Silesia).
The aim of these investigations is the identification o f correlations between the population number o f the towns studied and the assemblage of social-eco nomic features on the background o f regional differentiation and their change ability in time. These studies represent continuation and, at the same time, veri fication o f problems o f regional differentiation o f factors influencing popula tion transfonnations o f towns (Runge A., 2004b).
The analysis included the assemblage of 12 social-economic features assuming that in the studied period of 1977-2001 they influenced the most the size o f urban population (Domański, 1995). The features concerned the assemblages of demogra phic, infrastructural, financial and economic features. Some difficulties in obtaining a complete range of comparative data caused than in 1977 the investigations inclu ded 10 features (without the employment in trade per 1,000 inhabitants and the percentage of the employed per total employment), and in 1994 the investigations did not consider population per 1 shop. In order to solve the problem of the work, the method of step regression analysis was applied (Draper, Smith, 1973). The de gree of the adjustment of the significant features in the model of step regression was determined using a corrected coefficient of determination R2. The applied cor rected coefficient of determination was taken after Ratajczak (2002, p.81-82 ), who demonstrates that having large number of independent variables the use of the cor rected coefficient decreases a hazard of choosing false models.
The first stage o f investigations included the identification of these social-
economic variables, which in individual time intervals, appeared to be the most intensively correlated with the size o f population o f the historical area o f Silesia. Tab. 1 shows these variables, which were selected with the use of the method of step regression. Using the coefficient o f corrected determination, a degree o f ad justment o f the step regression model to input data was demonstrated. As far as the type and order o f the identified significant features in the model in the period o f central planned economy were in fact constant, in new social-economic reality they considerably changed. Instead of the so far leading role of trade, the services connected with health protection predominated (1994). This resulted from rapid privatisation o f trade and also from the shifting o f the considerable part o f trade activity to informal sector (street trade, market trade) in the first years of transfor mation period. On this background, a feature from the health protection
assem-CHANGEABILITY OF SOCIAL-ECONOMIC FACTORS IN THE DEVELOPMENT..
blage gained larger importance as this type of activity underwent privatisation in much smaller range. The symptom of developing qualitative changes was the origin of a financial feature in 2001. In the period of the already established structural changes o f political-economic transformation, the increase o f the num ber of significant changes in the regression model occurred. The changes in a hie rarchy of the significant features in the model of step regression should be explai ned as a derivative of political-economic transformations from the turn o f the 1980s and 1990s. In the conditions of centrally planned economy norms used in spatial management (local and regional) caused the necessity o f the adjustment o f number and structure of trade-service establishments to the size of population. These norms were not always fulfilled but in the field o f basic services for popu lation they were the closest to the declared requirements. Other problem was the relation between the degree of saturation of service establishments and a real fulfilment o f the inhabitant needs (these standards were very low as compared to European standards). New social-economic reality o f the 1990s caused that norms were left behind what influenced a development of demand-supply mechanisms in local and regional development. Therefore, among the variables which were the most correlated with population size occur since then both these, which were the remainders of “norm period” and those which are typical for transformation period (e.g. financial features).
Table 1. Social-economic features significantly correlated with changes o f population in Silesian towns in the period 1977-2001 (total)
Ye a r s So c ia l-e c o n o m icf e a t u r e ss ig n if ic a n tint h es t e pr e g r e s s io nm o d e l R 2
1 9 7 7
Pop ulation per 1 s ho p
Doctors per 1 0 thou san d population Natural increase per 1 ,0 0 0 population
2 8 .8 7
1 9 8 5
Po pula tion per 1 s ho p
D octors per 1 0 thou s an d population Natural increase per 1 ,0 0 0 population Incom es of counties budgets per 1 inhabitant E m p lo y e d in trade per 1 ,0 0 0 population
36 .0 9
1 9 9 4
D octors per 1 0 thou s an d population E m p lo y e d in trade per 1 ,0 0 0 population
Percentage of em plo yed in industry in a total em p lo ym ent E xp e ns es of counties budgets per 1 inhabitant E m p lo y e d in trade per 1 ,0 0 0 population
1 5 .4 6
20 0 1
Exp e n s e s of counties bu dgets per 1 inhabitant Doctors per 1 0 thou s an d population Pop ula tion per 1 s ho p
Em p lo y e d in trade per 1 ,0 0 0 population Pop ula tion per 1 ro om in dwellin g resources Coeffic ient of m igra tion efficiency D w e llin gs com ple te d per 1 ,0 0 0 population Natural increase per 1 ,0 0 0 population
3 1 . 1 1
A statistical adjustment of features in the regression model is not too high, but a trend o f the increase o f this adjustment is visible. This trend occurred first in the period 1977-1985 and then in the period 1994-2001. Weak adjustment of the model suggests large regional differentiation of factors influencing population trans formations. Thus, significant features correlative with the population size within individual settlement systems were analysed, which gave regional differences.
In the second stage o f investigations the answer to the question was investi
gated whether there are differences between historical-geographical parts o f Si lesia (Lowe Silesia, Opole Silesia, Upper Silesia) in the identification o f these variables, which correlate the most with the population size (Tab. 2).
Among the three areas studied, Lower Silesia shows the weakest adjust ment o f step regression models to the observed data, whereas Upper Silesia - the strongest. Generally the intensity o f this adjustment grows from the west to the east. Moreover, in the period 1977-1985 a clear step-like trends o f chan ges o f this adjustment is observed (except Opole Silesia), which was caused by strong social-economic perturbations connected with the political turn o f 1989. Therefore the trends o f the increase o f the adjustment o f significant features in the regression model is visible in the periods o f relative political-economic sta bilisation - in the years 1977-1985 and 1994-2001.
In Lower Silesia, among features significantly correlated with population size infrastructural features, mainly trade and health protection, predominated. Dwelling appeared in 1994. The economic feature occurred in 1977. Among demographic features the influence o f natural increase is observed in the 1970s and 1980s, and migration in 2001. A significant change in 2001 consisted in predomination o f a financial feature over infrastructural features.
The hierarchy of the identified variables for Lower Silesia appeared to be similar to the whole o f towns o f the historical area o f Silesia. The weak adjust ment o f regression models may be explained in two ways. Firstly, it results main ly from considerable differentiation o f the assemblage o f towns studied, both in dimensional and functional terms, and secondly, from their administrational at tachment to different provinces (at present to 2 but in previous administration division to as many as 7 provinces). Settlement system of Lower Silesia shows strong position o f Wroclaw in relation to other middle and small towns o f this area, which creates certain consequences in form o f “washing out” of develop ment factors - the effect o f the so called metropolis shadow. In the shadow of Wroclaw metropolis only small towns occur. The zone of medium and large towns occur in large distance o f Wroclaw. These towns are not considerably influenced by Wroclaw because they have their own, strong town-creating functions (many of them used to have administration functions on province level). This is reflec ted in conditions and differential o f demographic and economic changes o f the investigated assemblage o f towns (Wroclaw - demographic transformation).
T ab le 2. S o ci al -e co n o m ic fea tu re s sign ifica ntly co rre la te d w it h th e ch ang es of po pu la tio n in S il es ia n to w n s in th e p er io d 1 9 77 -2 00 1 (in re g io n a l sy st e m s)
CHANGEABILITY OF SOCIAL-ECONOMIC FACTORS IN THE DEVELOPMENT...
0 3. 2 =3 ” g . 5 £ k_ Я =3 О ) О) Q- Q. С Q- г-i о :— M O щ о о 5 оз i— о -а Ü Л Ц •§ iS 1 5 ~ 1 5 О Й - Û СО О Х — -5— чи w ° С с С - С 00 о •— с с | | СО 1— _СО =3 О) О "со о. с: О.— о — о ^ Q - CD CD. CD ^ CD CD -О CD CD с CD . (Л CD J - 1 —i <—> CD L " О 7Г CD. >3= CO ... OJ i2 = -s5 s, со а а > -СО О СО -О 03 D г Ш m а Я 00 5 аз ?з 8 ^ g ■§ - §.■£-§ a f # 1 3 О. £ ■ 8 Е iS 1— ÎDCD -CD CD ч— CU H— о со а со а а с £ ° СЗ XD — — с — er с: cd « с о £=оо"сэс= гт) i .— r i —. .— CD CD Р О ^ о о . Е Е з= с о с — с о -= - I f &О CL. f a 0 3 CD. ;= о ^ Q— to iS 8 -с г- = Я •- .E с Q _ Q . т_ ----2 о о i— ■-*=: о. Q- аз>— | о о о. « ’ О О 2-0 о_ L т— f— 03 02 . 1_ "О ^ ш 03 D S ; CL = -0 => l 'S È ’ S и 1 4-4 СО Х = .Я : .Я о о •*=; C D .TD О Щ " = 8 о _ с ^ « со -О со ° о э cd а з со С > 1 « аз = о с с 0 о. Я о 1 Е X с О Ш LLJ — О- о ° -2 О - Q 3 н— 02 0 3 2 аз .ЕР'О о аз SZ CD. со аз О О . с О § ^ 2 ~ аз J S .оз со аз о g со â= Е Е —’ 7=; q О. Я ш W Q. о F о У V о. ш ü S lii 1 л s ^ со о a a § ' 2 Q.TI » s -CD. О З -С= О СО о TD Ö D ° . ^ -° ^ "со l a i о 03-^ со о аз CD о со • о. y: сз оз S 8L о Я g î Я тз I* i Е 2 о 3 О = а. ) и ca Е ) £ z LU СО XD О о о О — 8 5 т: iS з g 3 Ql а CD. g W О CD. аз CD. CD CD3 ü 8 = g- О -Q ° СО L. со -С d аз аз 02 о Q.X3 1 £ аз g аз о со о о.-г- аз о с аз о 'S О CD. О
11|1
CD. о 2 Я О О СО Я а_ о z: -Е *- -В =з й аз о. cd. 0 . 2 о CD iS а Р -gcd ri CD i -w CO i— 1-L -(O CO 0 3 T D аз о о. cd 5 2 02 02 R 5 С О с to аз 1 -5 ° -i о. iS 1 О CD -Q. 0 3 s ■ 5 ? " 2 83 03 ё О О CD Ö аз о v£= — сд оз со “ о со Я 0 2 с Е 0 2 Я -S f i | g о о О — о Я - Э 2 О Р ^ ÜL LIÜ S - sсо .02 -с_з аз CD аз 2 Е оз Er аз о Q_аз о аз cd. s<0CX50xQ3>C2 u j üZ D u j û- QlD ■— .E E 8 L-JÜ3 л— =>, ) —i . c_ g co : .E S ■ 3=: —1 i CD. q_ , . X-ï> o 3 CD > > о з с -ö 11Я “ <= . o - o — оз o o ^2 >i <— <■" — CD. о. о CD CD О О CD. аз CD CD О - я m аз — Q. о iS Я XD CD.I CO> 1 3 CD2 ( XD CO 4 о о ' O TD о er Я* со = о С Ш п § я- о. Е Щ. ^ E X аз о _ о S.1-X аз о X Е LU Q_ Q_ LU LU аз Q- аз ^ E о О LU О О -° п о га о 8 -я - CD. О 3 CD -2-CD 2 Я 3 CD. О . . TD О < 0 Е 1 ri D CD.т Ä.O ОЗ ] Q а О о £ Л Н aMoreover, it should be remembered that Lower Silesia in the post-war pe riod showed relatively small investment expenditure as compared to other areas o f Poland (Ciok, 1993). Underdevelopment o f social infrastructure, especially in the southern part o f Lower Silesia (former Jelenia Góra Voivodeship and Wałbrzych Voivodeship), development o f Legnica-Głogów Copper Area in the northern part, or supremacy o f Wroclaw represent factors which influenced the most regional and local differentiation of features significantly correlative with population size. These factors spatially modify the correlations between the population size and the included social-economic variables.
In the towns o f Opole Silesia in the period o f central management o f economy (1977, 1985) the position o f the significant features was similar to the whole towns o f historical Silesia. Until the mid 1990s, infrastructural fe atures o f trade, health protection and dwelling predominated, however, still in the period o f centrally planned economy the spectrum o f these features inclu ded also an economic feature (the employment in industry per 1,000 inhabi tants) and financial features (incomes and expenses o f counties budgets per
1,000 inhabitants). There was the lack o f the influences o f demographic featu res. In 1985, the coefficient o f migration efficiency became more important. Later, the variable connected with dwelling situation and trade predominated (1994) followed by variable connected with financial and economic sphere (2001). Despite generally difficult situation o f housing industry in the 1990s, the individual achievements in that sphere in Opole Silesia place this region in better position than other parts o f Silesia. In 2001, similarly to other regio nal systems, the financial feature predominates with large contribution o f in frastructural features (trade, dwelling), but unlikely to Upper or Lower Sile sia, the role o f economic features considerably increases. Large position o f the employed in industry suggests that the role o f traditional economic activi ty in Opole Silesia is significant, which is especially connected with its eastern part adjacent to Upper Silesia. The influence o f demographic features was the weakest (migrations). The outlined trends differ from a general trend visible in the scale o f the whole towns o f Silesia.
The settlement system o f Opole Silesia is strongly integrated by the influ ence o f Opole, what creates considerable homogeneity o f features influencing demographic changes. Therefore this area shows strong adjustment o f the signi ficant features in the regression model.
At first, in the towns o f Upper Silesia, similarly to other parts o f Silesia, infrastructural features correlated the strongest with the population size. Also financial features were o f significant importance (like in Opole Silesia). Dif ferently to other regional systems studied, demographic features, especially migrations, played a significant role influencing population changes. A featu re reflecting the importance o f industrialisation processes occurred only twice (1985, 2001) in the period connected with the drop o f the number o f the em
CHANGEABILITY OF SOCIAL-ECONOMIC FACTORS IN THE DEVELOPMENT..
ployed in industry. The state o f saturation o f the job market with workforce (restraint o f the demand for workforce) in the mid 1980s was connected with the termination o f large industrial enterprises in the region (Runge, K łosow ski, 2000; Runge et a l , 2003). The restrained reduction o f the employment in industry intensified at the end o f the 1990s. Wide spectrum o f the significant features and also very high level o f the adjustment o f the models o f step re gression to output data suggest that the areas o f domination o f large or me dium towns (or smooth settlement structure) show better adjustment in this range than strongly diversified areas (with domination o f one centre) or with large percentage o f small towns.
This conclusion requires however a thorough verification. Therefore the
third stage o f statistical investigations was carried out. The whole assemblage
o f Silesian towns was divided into three classes according to their sizes: small towns (up to 20 thousand inhabitants), medium towns (20-100 thousand) and large towns (over 100 thousand). Next, for the assemblage o f each size-class, the models o f step regression were constmcted, which made it possible to iden tify the features that correlate the strongest with population size (Tab. 3).
The results o f this investigation confirm the thesis that the most significant correlative features concern most o f all large towns, then medium towns, whe reas small towns show weak correlations. As far as in the period o f centrally governed economy, in large towns, the intensity o f the model adjustment was increasing almost reaching 100%, in medium towns it rapidly decreased in the end o f the period o f planned economy, which suggests that towns o f this size were sensitive to the growing symptoms o f economic crisis. In small towns the level o f the features adjustment was constantly low.
In the new political conditions, large towns, after initial rapid correlation weakening, showed its significant increase. Similar trends were observed in medium towns, however in this case the weakening o f correlation in regression model was very strong, which was conditioned by perturbations connected with the establishment o f new political-economic system. After the period o f sponta neous social-economic transformations, a gradual restoration o f the correlations studied occurred, which was accompanied by considerable enlargement o f the spectrum o f significant features in the model. The strongest adjustment o f fe atures occurred in 1977, and then it gradually decreased - first as a result o f growing economic crisis in the beginning o f the 1980s, and next because o f a very difficult beginning o f political-economic transformation (1994). A stabi lisation o f political-economic transformations occurred in the increase o f the adjustment o f the features studied n 2001. Medium towns appeared to be there fore the most sensitive to political-economic transformations o f the turn o f the 1980s and 1990s.
In small towns further mitigation o f already weak correlations in the re gression model occurred.
Ta bl e 3. S oc ia l-ec on om ic ch an ge s of si gn if ic an tly co rr el at ed w ith th e po pu la tio n ch an ge s in Si le si an to w ns in th e pe rio d 19 77 -2 00 1 ac co rd in g to th e siz e cl as se s of to w ns о о to to = о. g. O Q O Q . Ю . -5 O O Я CD CD Й-S O L 03 i %'S g i g • — Q . " O Ш C L О О О О o a. Q i i s -8 га a c l о О е= Cl г Ç L .аз со L о С Э J о о £ .02 L с д о --Р ~ g ’cz О - g о 1 с: ÇZ Я о : vi О о о о Я ° — 4 _ ° •‘Г О С W ° ( л œ С П СО о Ü Е œ W 1 1 | ! О Q — UJ Ë •-"ö-E Q- Q. Я — i S -f i l O - Q . о О Я 0 о _ о ^ СО V— Q - з W Щ с 5 =3 О . О : О 2 о ! * = а ! *=: со - О ~ с о - о = С Cl 03 CD . Ï И ^ .Я "о ч— " О î-t— 0 3 Я . Е ® * Е - а Е о S ■ £ г ; (О о iS -j£ & ^ g с я И Е О =£ - и о О g с о Cl c d £ Д Ш О о. р Е Е ^ о* с о fc 03 О l u Q _ g.g-1 I i £ o 2 2 '0 cr> 3 c p CD ? О CD s. : аз Я i яг 0 : o i О О "C3 CZ “ ~ C D ----О CO я я-О я-О - p -'iü Q-- E CD О . o CD CD. ° ° 8 O CD И ^ CD ! 2o .E о L.E - .E _g i -o Q .-0 о i >, CO O o o tg L 0 - 0 g - оз . L U Q L U ^ 0 3 - r - .j= f E to to 'to о О . с 03 0 . 2 ^ Я . cd ■ o T - i E .E fe ® > co g - “ 'S S O C L % B-E Я — Ł_L 2 E o a5 ^ O - w CD 3 o . _ Я 2 CD g -= "g E V СО СЭ > D "Й O O o . - = ; CD o o 5 o .52 D - o r - ' CD CD Q - 1_ — O 0 03 «= 2.0 0- 1- O c-™ O c Q-Й ~ ~ O T“’ 00 2 а аз -§ ~ ” S’— : » ss-E o "03 — . P - CD o to to ^;p— - CD 9 2 ^ «4 /Л “tw П "iCD Ю CO o. o. g - g O O E CD Q _ Q _ LLJ O Ó -— с — F О о -Я а= S Z t D -О . О ) E CD I LU Q_ 8 "e .lCO CD^ со аз E — - > , со о Lf^ l_ — —■ 92 оз. о “ со О ^ ^ »2 Q-g о. о 2 0 5 0 9- g- : ■q о g i о . Я °* < ! ° - § î 1 СО т з а з .; § î Я CD CD i I t S i ^ О (— ( fe 0 “ i O . CO - O ( 0 3 CD •■ E o: 0= 0 ! o §2 g-j o ^ E < O O L U Z
CHANGEABILITY OF SOCIAL-ECONOMIC FACTORS IN THE DEVELOPMENT.. _ Q CX to ■ o -5 i О - К CD О О CL 0.0 о о 0 3 Я а5 ^ т~ i â g CD ; = СХ0 5 а З — *-* о оз С С О О — •+=; о -о -а ° с: а з о с о — >-% cd г = о о Е g а а о 5 Е Е ^ о ^ Л ~ -Ш О = §2 °-а5 . * о -о ; с <=> = Е о -О CD 5 О ex g CD *— " " О , »_ g 03 CD Ч e x с : c l с с ~ с .g lg .2 со >> СО ä g - ä о Е о ~ . lu а_ 0 3 CD Е ^ • а аз а 5 О С О 2 о а з ° - з о 034— 33 CD - о >-3 “ О CD 3 _Q Cl о • ,дз л Е •— с o' Ö a= аз -2 Е с о с о -Q о • Е 00 * g CD to l j e x l_< О CD £Е ,=• Q _ LU û _ О Q _ CD - g P- 8 .È? s." ■=; О | S CX - г - О Q _ CO ° * O « T 3 r - =3 CD W о Ш T— - w Q _ ® ° § - 8 § S.Ä: c/30 3 о = X о g 3 о ^ . Q О О ’S Я <D Л н
It should be concluded therefore that the selected for the investigation social- economic factors correlated in the first case with population changes in large towns, showing slight mitigation o f this relation in the beginning of the 1990s. In case o f medium towns, first the decrease of their importance was observed, and then in new political-economic reality - the increase of their position. In small towns - independently to the period o f investigations - the discussed social-eco nomic factors correlated less and less with population changes.
Simultaneously in the area o f historical Silesia, three trends o f correlations between population and the selected social-economic factors occur. They are as follow:
— in the scale o f the whole territory these correlations are rather insignificant but growing - first until the mid 1980s and then after a break in 1994 still growing;
— in the regional pattern there is an increase of correlations from the west to east;
— in the pattern o f size classes o f towns these interdependencies correlates to individual size groups o f towns - the strongest occur in large towns, we aker in medium towns, and the weakest occur in small towns (they get smaller together with the decrease o f population number).
Another problem is a hierarchy of features in the regression model in indivi dual size groups of towns. In large towns a large spectrum o f features occurs which are significantly correlated with population size in all investigated time cross-sections (Tab. 3). In 1977, this concerned 6 features mainly infrastructural (health protection, dwelling, trade) - therefore, similarly to regional patterns, but also economic and demographic features occured. In 1985 the number o f features increased to 10 and the appearing financial features, accompanied by infrastruc tural features showed the predominating influence on population transformations. Also economic features occurred as well as the whole spectrum of demographic features. In new political-economic conditions, the adjustment o f features signifi cantly correlated to population changes slightly decreased but it was still strong. However, the structure o f features significantly influencing population changes “retarded” - and was similar to that from the 1970s (domination o f infrastructural features over demographic feature). After overcoming perturbations connected with political-economic transformation, a subsequent strong increase o f signifi cance level o f the features studied occurred and some changes in their hierarchy took place. Financial features on the first place followed by infrastructural featu res were the strongest correlated with population size.
In the beginning (1977) in medium towns, trade was the strongest correla ted with size population, but in fact, population transformations which occurred here were influenced by demographic features. A financial feature also occur red. In next time cross-sections, the influence o f demographic and financial features decreased. Until 1994, on the background o f permanent decrease o f the
CHANGEABILITY OF SOCIAL-ECONOMIC FACTORS IN THE DEVELOPMENT..
correlation o f the investigated features in the regression model, there was a do mination o f infrastructural features supported by economic feature and demo graphic feature. The tightening o f correlations between population changes and the assemblage o f the investigated features occurred in the last time cross-sec tion, after the establishment o f political-economic transformations. The increase o f the significance level was accompanied by strong widening o f a spectrum o f significant features in the regression model. However still, infrastructural featu res supported by other groups o f features predominate.
In the 1970s, population changes o f small towns were mainly influenced by demographic feature (coefficient o f migration efficiency) as well as the assem blage o f infrastructural features which were supported by the impact o f indu strialisation. In the 1980s, spectrum o f significant features enlarged and became differentiated. Infrastructural features predominated - similarly to most o f the systems studied, supported by economic and demographic features. In the con ditions o f transformation, economic features strongly supported by infrastructu ral and financial features (1994) appeared to be basic features influencing popu lation changes. In the last time cross-section, only 4 significant features influen ced population transformations in small towns, but their adjustment in the re gression model is rather weak. Still economic feature predominated.
It may be stated therefore that large towns due to their spatial-functional and social-economic complexity show a complex pattem o f correlations betwe en population size and social-economic conditions. This results in a wide spec trum o f factors showing the correlation with population size and influencing the direction and dynamics o f population changes.
Middle towns in many cases represent centres which underwent considera ble urbanisation after 1945, connected with location in these towns certain ty pes o f economic activity. This was accompanied by the development o f house building (together with social infrastructure). The relation o f these three consti tuents was especially visible in the 1970s, so it should be assumed that large value o f R2 for medium towns in 1977 results from this fact. Because many medium towns represent centres o f monofunctional character, therefore such factors as difficulties on job market or decline o f house building contributed to considerable decrease o f the role of the investigated factors in this group o f towns. In the beginning o f the 21st century, an increase o f the adjustment be tween the variables studied is visible, which should be explained by larger de termination o f local authorities in searching for new development factors (Pary sek, 1995).
In the period o f centrally planned economy, population processes occurring in the studied settlement systems in general were stimulated by similar growth factors (Runge, 2004b). Development o f territorial governments and political- economic changes o f the transformation period revealed regional differentiation o f factors influencing social-economic development o f these systems and gave
them chance for individual development according to their regional conditions and settlement patterns.
CONCLUSIONS
In the period 1977-2001, the assemblage o f features significantly correla ted with population size o f Silesian towns underwent considerable changes. In the scope o f regional differences, the most important include:
— similarities o f features significantly correlated with population size o f towns in all the investigated regional systems in centrally managed economy, and then differentiation o f this spectrum o f features in new political-economic conditions;
— transition from the domination o f infrastructural features in the regression model to financial features supported by various groups o f features in indi vidual regional systems;
— clear differences in the adjustment intensity o f the significant features in the regression model and in the trend o f its transformations - the weakest adjustment with a trend o f further mitigation o f these correlations in Lower Silesia, with strong and constantly increasing adjustment o f these correla tions in Upper Silesia and Opole Silesia. Simultaneously, in all the regional systems an extremely significant decrease o f these correlations occurred in the period o f transformation conversion (1994), followed by a renewed in crease o f these correlation in Upper Silesia and Opole Silesia, whereas in Lower Silesia considerable differentiation occurred resulting in mitigation o f the analysed relations (lack o f statistical significance).
In the scope o f differences between the size groups o f the towns studied, the most important include:
— the occurrence o f strong correlations between the assemblage o f the inve stigated social-economic changes and population size in the group o f large towns, weaker correlations in the group o f medium towns and very weak correlations in the group o f small towns;
— the occurrence of similar trends of changes in the adjustment o f the significant features in the regression model like in regional systems - i.e. these size cate gories o f towns which in the time o f centrally governed economy already sho wed large significance o f features in the regression model (large and medium towns), after the period o f perturbations connected with transformational co nversion, renewed these correlations, whereas in small towns these process became mitigated and these correlations became statistically insignificant; — structural changes o f the significant features in the model consisted in the
domination o f infrastructural features by financial features in large towns, and domination o f demographic features by infrastructural features in
me-CHANGEABILITY OF SOCIAL-ECONOMIC FACTORS IN THE DEVELOPMENT..
dium towns. The appearing in new political-economic conditions domina tion o f financial features clearly observed in all the regional systems is not therefore so clear in individual size groups o f towns.
The differentiation o f the adjustment intensity o f these features between the individual size groups results from larger vulnerability o f small towns to extre mely differentiated local and regional conditions, whereas medium towns in their development are more intensively influenced by regional differentiations than local differentiations, and large towns side by side with regional conditions are influenced by factors o f national level. The vulnerability to the analysed spectrum o f factors depends also on the functional structure o f towns - large towns o f differentiated functional structure react on the influencing factors in more flexible way, whereas, taking into account large functional specialisation o f small towns it is difficult to determine the assemblage o f growth factors which would be identical for so differentiated set o f towns. The differentiation o f factors and lack o f their adjustment in the regression model probably sugge sts the increase o f local conditions over external influences in social-economic transformations o f these towns.
REFERENCES
Bagdziński, S.L. and Maik, W. 1995: Uwarunkowania polityki regionalnej i przestrzen
nej w okresie transformacji systemowej. In Bagdziński, S.L, and Maik, W, and Potoczek, A, editors, Polityka rozwoju regionalnego i lokalnego w okresie trans
formacji systemowej, UMK Tomń, pp. 9-34.
Ciok, S. 1993: Zagrożenia rozwoju społeczno-gospodarczego Sudetów i możliwości
ich eliminacji. In Loboda, J, editor, Przestrzenne problemy rozwoju społeczno-go spodarczego Sudetów, Studia Geograficzne LVIII, Wydawnictwo Uniwersytetu Wrocławskiego, Wrocław, pp. 187-198.
Draper, N.R. and Smith H. 1973: Analiza regresji stosowana, PWN, Warszawa. Drobek, W. 1999: Rola miast zdegradowanych w sieci osadniczej Śląska, Instytut Ślą
ski, Opole.
Domański, R. 1995: Tematyczna koncepcja statystyki miast. In Kosiedowski, W, editor,
Gospodarka przestrzenna i regionalna w trakcie przemian, UMK Toruń, pp. 34-45.
Parysek, J.J. 1995: Rola samorządu terytorialnego w rozwoju lokalnym. In: Parysek
J.J, editor, Rozwój lokalny: zagospodarowanie przestrzenne i nisze atrakcyjności
gospodarczej, Studia KPZK PAN, t. CIV, PWN Warszawa; pp. 37-62.
Ratajczak, W. 2002: Nierozwiązane problemy analizy regresji w badaniach geograficz
nych. In Rogacki, H, editor, Możliwości i ograniczenia zastosowań metod badaw
czych w geografii społeczno-ekonomicznej i gospodarce przestrzennej, Bogucki
Wydawnictwo Naukowe, Poznań, pp. 75-86.
Runge, A. 2004a: Miasta średnie w sieci osadniczej Śląska. In Słodczyk, J, and Raj-
chel, D, editors, Przemiany demograficzne i jakość życia ludności miast, Uniwersy tet Opolski, Opole, pp. 85-100.
Runge, A. 2004b: Typologia dynamiczna miast Śląska w zakresie przemian społeczno- gospodarczych. In Jakubowicz, E, and Raczyk, A, Przekształcenia regionalnych
struktur funkcjonalno-przestrzennych t, VIII/2 - Regionalny wymiar integracji eu ropejskiej, Uniwersytet Wrocławski, Instytut Geografii i Rozwoju Regionalnego,
Wrocław, pp. 103-114.
Runge, J. and Kłosowski, F. 2000: Migracje ludności miast województwa śląskiego w latach 1977-1997. In Szymańska, D, editor, Procesy i formy ruchliwości prze
strzennej ludności w okresie przemian ustrojowych, UMK Toruń, pp. 121-134.
Runge, J. and Kłosowski, F. and Runge, A. 2003: Conditions and trends of social- economic changes of Katowice region. Bulletin o f Geography; socio-economic se
ries, no. 2/2003, Nicolaus Copernicus University, Toruń, ss. 85-102.
CORRESPONDENCE TO:
Anna Runge, Section of Social Geography, Department of Economic Geography, Faculty of Earth Sciences, University of Silesia,
Będzińska 60, 41-200 Sosnowiec, Poland [e-mail: jrunge@ultra.cto.us.edu.pl]
