Residential Preferences and Population Distribution

DEMOGRAPHV@

Volume 12, Number 3

August 1975

RESIDENTIAL PREFERENCES AND POPULATION DISTRIBUTION Glenn V. Fuguitt Department of Rural Sociology, University of Wisconsin, Madison, Wisconsin 53706

James J. Zuiches Department of Sociology, Michigan State University, East Lansing, Michigan 48824

Abstract-Public opinion research has revealed decided preferences for living in rural areas and small towns, and proponents of population deconcentration have interpreted this as support for their policies. This study, based on a national sample, yielded similar results, but when we introduced the additional possibility of a preference for proximity to a larger city, then the rural areas preferred were found, for most respondents, to be those within the commuting range - of a metropolitan central city. Although persons wishing to live near large cities were found to be looking for the same qualities of living sought by those who prefer a more remote location, these findings are not, in general, consistent with the argument that public preferences support strategies of population dispersal into nonmetropolitan areas. Instead they indicate that most of those who wish to live in a different location favor the peripheral metropolitan ring areas , that have, in fact, been growing rapidly by in-migration. INTRODUCTION Recent social trends relating to the decline of central cities, urban sprawl, rural depopulation, and disparities among communities have served to focus attention on the need for population distribution policies in the United States (see, for example, Ad Hoc Subcommittee on Urban Growth, 1969; Advisory Commission on Intergovernmental Relations, 1968; Beale, 1972; Commission on Population Growth and the American Future, 1972; Fuguitt, 1971; Hansen, 1970; Morrison, 1970; National Goals Research Staff, 1970; President's National Advisory Commission on Rural Poverty, 1967). Specific recommendations have included proposals to develop depressed rural areas, to encourage migration to middlesized cities, and to plan for more orderly urban expansion. Various strategies also have been suggested to redistribute the population away from metropolitan areas 491

and to retain population in nonmetropolitan areas. An important element figuring in this discussion is concern about public preferences and attitudes on desirable places to live. A policy that provides eommunity and housing options compatible with preferences should have a greater chance of success and could be expected to lessen any discrepancy between the actual and ideal distribution of the population. That such a discrepancy exists is one argument used by proponents of population deconcentration. For more than a decade, national public opinion surveys have found a, decided preference for living in small towns and rural areas as contrasted to larger cities. Thus the coincidence of public interest and private preference is pointed to as a positive reason for a population dispersal policy since, "except for the economic pressures, many city dwellers would eagerly move

492

DEMOGRAPHY, volume 12, number 3, August 1975

to the country" (National Goals Research Staff, 1970, p. 54). Despite this stated dissatisfaction with large cities, there has been no major exodus to medium-sized cities, small towns and rural areas of nonmetropolitan counties. In fact, during the 1950's over 1,400 nonmetropolitan counties experienced such heavy out-migration that they declined in population, and about 1,300 nonmetropolitan counties declined in the 1960's. Since 1960 a number of formerly declining nonmetropolitan areas have begun to grow (see Hansen, 1973; Beale, 1974), and since 1970 a net migration gain has been reported for the nonmetropolitan sector as a whole. Nevertheless, Current Population Survey results (U.S. Bureau of the Census, 1974) indicate that between 1970 and 1973 the rate of net migration gain for metropolitan areas outside central cities was three times the rate for the nonmetropolitan United States. In this paper, we offer evidence that this paradox between expressed public opinion and actual migration can be explained at least in part by the way residential preference questions have been asked in earlier surveys. Whereas previous studies have generally consideredonly the size of place in which respondents prefer to reside, we have further distinguished places according to proximity to large cities. Our initial work (Zuiches and Fuguitt, 1972) was based on a survey of Wisconsin residents. This paper reports the results of a nationwide sample survey which allow us to draw conclusions relating to preferences by city size and proximity to a metropolitan center for the adult population of the continental United States. PREVIOUS RESEARCH ON RESIDENTIAL PREJ:;'ERENCES

Some previous results on preferences by size of community are reported and cited in Table 1. Because of differences in the questions asked, these distribu-

tions are not precisely comparable; yet the findings are broadly similar, showing that (1) most respondents would like to live in small towns and rural areas and (2) the proportion having this preference exceeds that currently residing there. All of these studies, however, have a common limitation: they fail to distinguish certain key relational characteristics of places people claim to prefer. Among these, nearness to a large city may be of paramount importance. Failure to distinguish proximity differences could well mix, on the one hand, the respondent who prefers the small town or country milieu of, say, Philadelphia's suburban Main Line with, on the other hand, the respondent who has a small town in central Nebraska in mind. The same "small town" descriptor on the survey can have quite different meanings to different respondents. This potentially misleading effect was first examined in our statewide survey of Wisconsin in 1971. After an initial sizeof-place question, we asked respondents not preferring a large city whether or not they would prefer to live within 30 miles of a large city. Judging from the initial question, our results broadly agreed with those of Gallup, the Population Commission, and others. The distance-qualifying question, however, showed that well over one-half of the 79 percent preferring to live in small towns and rural areas stated that they would like their residence to be within 30 miles of a city over 50,000 in size (compare last two columns of Table 1). This finding shed considerable light on the paradox noted above: the realization of these preferences would result in a net movement out of central cities and nonmetropolitan areas and into metropolitan rings, which is consistent with the overall trends that prevailed in Wisconsin at least through 1970 (Zuiches and Fuguitt, 1972). A study by Dillman and Dobash (1972) done at about the same time in Washington State also casts doubt on the

1

0

100 100

1

29 27b

25

18%

100

1

31 b 24

18% 26

100

1

13% 31 32 23b

1912

104

1

25 54c

6% 18

National lUldlife Federation 1969

100

5

31 b l6

27

21%

Harris 1970

100

1

21

35

100

0

28%e

11

l8d d 30 d 34 44e

10%e

14

l6%e

Zuiches-Fuguitt (Wisconsin) 1971 Within Beyond 30 30 Total mles mles

17%d

Population Commission 1971

Sources: 1948 Roper study cited in Lee et al., 1971, P. 33; G~llup results in American Institute of Public Opinion, 1970, and National Area Development Institute, 1973; National Wildlife Federation survey Teportell in National Wildlife Federation, 1969; Harris survey in Louis Harris and Associates, Inc., 1970; Population Commission results from internal. memorandum of The Population Commission, 1971; Zuiches-Fuguitt from Zuiches and Fuguitt, 1972.

a- Includes respondents preferring small cities and small towns. The Roper question was: "If you had a choice, where would you like best to live--in the country, a smali town, a small city, a suburb, a large city?" b- Farm used instead of rural area in the Gallup question, which was: "If you could.live anywhere in the U.S. that you wanted to, would you prefer a city, suburban area, small town or farm?" Similarly, Harris question was: "If you could find just what you wanted in the way of a place to live and didn't have to worry a!:>out where you worked, would you want to live in a city, a suburb, a small town away from a city, on a farm, or where?" c- Includes respondents preferring rural areas (30 percent), mountains (15 percent) and seashore (9 percent). The National Wildlife Federation Survey permitted multiple responses and had six possible choices. d- "Large and medium" cities combined; suburbs of "large" and "medium" cities combined; small cities and small towns combined; and farm and open country combined in this table. The Population Commission question was: "Where would you prefer to live? On a farm, open country (not on a farm), in a small town, in a small city, in a medium size city, in a large city, in a suburb of a medium size city, in a suburb of a large city?" e- Medium-sized city and small city or town combined. Question was: "If you could live in any size community you wanted to, would you prefer to live in a large city of 50,000 or over, in a suburb next to a large city, in a medium-sized city, in a small city or town, or in a rural area." For those givip.8 one of the last three options, this was followedfby: "Would you preier to llve within cOll1lllUt:l.ng distance - say with:1.n 30 miles - of a large city, or be farther away rom a laree city?"

Total

100

31 lSb

Small towns Rural areas No opinion, other

22% 28

15% 20 a 4l 24

Cities puburbs

1966

Roper 1948

Gallup 1968 1970

I.-A Comparison of Surveys of Residential Preferences in the United States from 1948 to 1972

Preferred Residence

TABLE

ir

iii

-g'

~

C

I:T

i::::!.

e

::I

o

l:!:

lc

6-

III

II

::I

;

~

"U

!!:

::I

-

a: Cll

III

DEMOGRAPHY, volume 12, number 3, August 1975

494

interpretation of earlier surveys that a high proportion of citizens would prefer to live in small towns and rural areas away from large cities. Since their questions were worded quite differently, the results are not shown in Table 1. QUESTIONS AND SAMPLE

The data reported here are from a nationwide study building on our earlier work for Wisconsin. Basic questions very similar to those included in the Wisconsin 1971 survey were included in NORC's Amalgam Survey of the total noninstitutional U.S. population 18 and over conducted in November 1972. The questions were worded as follows: "First, we are interested in the kind of community you would prefer to live in now, if you had your choice. 1. In terms of size, if you could live in any size community you wanted, which one of these would you like best? A large metropolitan city (over 500,000 in population) A medium-sized city (50,000 to 500,000 in population) A smaller city (10,000 to 50,000 in population) (ASK A) A town or village (under 10,000 in population) (ASK A) In the country, outside of any city or village (ASK A) Don't know A. IF SMALLER THAN MEDIUMSIZED CITY: In terms of location, would you like

that place to be within 30 miles of a large or medium-sized city, or would you rather be farther away from such a city? Within 30 miles Farther away Don't know/doesn't matter" Similar questions yielded classifications with the same categories for present residence and residence at time of

birth. In addition we obtained information on the respondent's view of specific aspects of communities as related to his preferences and a ranking of preferred locations. These were analyzed giving particular attention to the relation between actual and preferred residence. Those interviewed appeared to have little difficulty responding to specific questions on distance from a large city and size of place. To get an impression of how accurately respondents could determine whether or not they live within 30 miles of a city over 50,000, the sample segments, usually census tracts, were classified by distance edge-to-edge to places of that size. We found that 64 respondents who were in PSU's more than 40 miles from a large city reported being within 30 miles of such a place, and 23 respondents who were in PSU's less than 20 miles from a large city reported living more than 30 miles away. These 87 were less than 6 percent of the total number interviewed, so we concluded that estimating distance is not a serious problem. Similarly, we compared reported present location with a classification of PSU's by size of place and found only 14 percent of the respondents differed by more than one class interval. Moreover, inspection shows that most of these deviant responses were plausible as, for example, some who reported being in large cities appeared to be in rural territory adjacent to a large city according to the PSU designation based on the 1970 census. The NORC sample is a multi-stage area probability sample down to the block level, where quota sampling is used. The primary sampling units are SMSA's and nonmetropolitan counties. Because of the quota feature, tests of significance are not appropriate. For details of the sampling procedure, see King and Richards (1972). RESIDENCE AND PREFERENCES

In Table 2 actual residence is com-

Residential Preferences and Population Distribution TABLE

495

2.-Actual and Preferred Residence of Respondents by Size of Place and Location with Respect to a Large City, United States, 1972

Type of Location City over 500,000 City 50,000 - 500,000 Subtotal Within 30 miles of city over 50,000 City 10,000 - 50,000 Places under 10,000 Rural area Subtotal

Current Residence

Preferred Residence a

20% 24

9% 16 25

44

11

16 15 24

34

55 6 4 9

Subtotal

6 6 9 21

19

Not ascertained

1

1

Total Number of cases

100 1,481

100 1,481

More than 30 miles from city over 50,000 City 10,000 - 50,000 Places under 10,000 Rural area

13 10

a- Forty-eight persons who responded "don't know'or "doesn't matter" to living Within 30 miles of a city were given a distance classification consistent with their current residence in this and succeeding tables. Source: National Opinion Research Center Amalgsm Survey conducted in November, 1972.

pared with preferred residence according Our results accord well with previous to both size of place and proximity to a survey findings (shown in Table 1) if large city. Almost one-half of the re- proximity to a large city is ignored. spondents in this nationwide survey re- The Gallup and Harris surveys of 1970 ported living in cities of over 50,000 and 1972, for example, show 55 and 47 population, one-third within 30 miles of percent, respectively, wanting to live in a city of 50,000 or more, and 20 percent small towns or on farms. The Populain more distant locations. If everyone tion Commission study and both our were to live in the location that he pre- Wisconsin and U. S. surveys show alferred, however, the distribution of popu- most two-thirds to three-fourths of the lation would look somewhat different. respondents preferring small cities and Under those conditions only one-quarter towns or rural areas. would live in large cities, and over oneOur data, however, reveal a key relahalf would be in easy commuting dis- tional prerequisite for desiring to live tance of such places. The proportion in small towns and rural areas: Table 2 living more distant, however, would re- shows that only nine percent of all remain almost the same. The largest drop spondents in the nationwide survey would in percentage, were such a shift to come prefer to live in rural areas, and another about, would be for cities over 500,000, ten percent in small and medium-sized and the largest gain would be for rural towns, if these locales were more than areas near large cities. 30 miles from a large city. (Comparable

496

DEMOGRAPHY, volume 12, number 3, August 1975

percentages for Wisconsin are 14 and 16.) Many people respond positively to the idea of rural living, but not where it would entail disengagement from the metropolitan complex. This suggests a clear desire to have the best of both environments-which may include proximity to metropolitan employment, services, schools and facilities, along with the advantages of the smaller local-residential community for familial and neighborhood activities. A complete cross-classification of respondents by their reported current and preferred residence is given in Table 3. The percentages in the principal diagonal represent those respondents whose actual and preferred types of residence coincide. This group includes 42 percent of the persons interviewed; the other 58 percent preferred a setting that differs from their current location. In only two residence categories (places under 10,000 within 30 miles and rural areas within 30 miles) do current and preferred residences coincide for more than one-half the respondents. For communities within the 30-mile zone, there is also a systematic inverse relationship between size of residence and preference for this residence. Only 36 percent of the residents in cities over 500,000 selected this type as their preferred location, but 67 percent of near (less than 30 miles) rural residents selected near rural areas as preferred. The least "popular" locations, as measured here, are large cities and medium and small towns away from a large center. Diagonal values for these residence types are 36, 35 and 21, respectively. This is true also for the Wisconsin study, with diagonal values, respectively, of 28, 22, and 33. Most of the differences between actual residence and preferences indicate a desire for smaller and/or more remote locations. About 40 percent of the responses are above and to the right of the diagonal, and only 17 percent below and to the left. Overall, however, the

data show the predominance of suburban and fringe-type locations in residential preferences. These are the categories with the highest agreement between residence and preferences and the categories most preferred by people currently living elsewhere. This generalization is shown clearly by combining the categories of residence and preference into three groups: cities over 50,000; smaller places and rural areas near large cities; and smaller places and rural areas away from large cities. Of people currently living in cities over 50,000, 52 percent would prefer smaller places. Conversely, of people not currently living in large cities, only eight percent would prefer to do so. More than three-fourths of those now living in smaller places near large cities would prefer this type of location, as would 42 percent of those not now living there. Of those more than 30 miles from a large city, over one-half preferred living nearer. But only 13 percent of the people living near or in a large city chose a more remote location. Table 4 is the cross-classification of current and preferred residences using this trichotomy of residence types. The remainder of the analysis is based on data grouped in this way, with the three categories refet'red to as "big city," "near" and "away," for convenience. (Whether a city of 50,000 is indeed big depends, of course, on one's viewpoint.) ORIGINS AND RESIDENTIAL PREFERENCES

Is a respondent's residential preference related to where he grew up or formerly lived? Table 5 shows the relationships between the size and location of birthplaces (as reported by respondents) and preferred residence and current residence for the three location types. Comparison of origin with current residence indicates the movement from more remote areas into big cities and their peripheries which has taken place over the last generation. If the preferred residences were realized,

3 2 1

City 50,000 - 500.000

Within 30 miles of city over 50.000 City 10,000 - 50,000 Places tinder 10,000 Rural area

16 7 1

7 4 3

41

15%

City 50.000500,000

15 4 1

43 10 7

17

l4r.

3 27 9

51 4

13

8

12%

21 37

11

14 23 67

20

11%

Preferred Within 30 Miles of City over 50,000 City Place 10.000- under Rural 50,000 10,000 Area

35 6 1

7 1 1

6

3%

City 10.00050,000

7 21 2

3 6 2

2

4%

Place under 10,000

14 45

8

8 3 14

2

5%

Rural Area

Residence More than 30 Miles from City over 50.000

3.-Preferred Residence by Current Residence, United States, 1972"

a- This table excludes 25 respondents who did not answer the question on current residence. Source: National Opinion Research Center Amalgam Survey conducted in November. 1972.

2

5

4

City over 500,000

a

36%

Current Residence

More than 30 miles from city over 50,000 City 10,000 - 50,000 Places under 10.000 Rural area

City over 500.000

TABLE

2

a a

100 100 100

100 100 100

100

a

2 0 1

291

100r. 0%

94 126

88

194 151 159

353

Number

Total

Not Ascertained

:lll CD

...

$"'ol

:::l

g-

c

c:r

~.

2

III

:::l

e. 0

i»

'C

~ c:

III :::l Q.

~

n

:::l

Cil

ir

Cil

'V

!!.

e.

:::l

a:CD

III

DEMOGRAPHY, volume 12, number 3, August 1975

498 TABLE

4.-Preferred Residence by Current Residence Trichotomized, United States, 197za Preferred Residence

Current Residence

Other, Less Than 30 City over Miles from Number 50,000 City

Other, More Than 30 Miles from City

Total

City over 50,000

641

48%

41%

11%

100

Other, less than 30 miles from city

500

7

78

15

100

Other, more than 30 miles from city

306

9

44

47

100

1,447

25

55

20

100

Total

a- Thirty-four cases were dropped due to "don't know" or "no response" to the preference or the residence question. Source: National Opinion Research Center Amalgam Survey conducted in November, 1972.

however, there would be a reverse shift out of cities, so that the proportion of residents in large cities and in rural areas would be less than the proportion born there; whereas the periphery of large cities would again capture an increasing proportionate share. Next we considered the association of place of origin as well as current residence with the preferences given for each of the three general residence types. Each cell of Table 6 gives the percent of people in the current residence and origin indicated by the row, who prefer the residence type indicated by the column heading. By comparing the first two TABLE 5.~Proportionof

rows, and the third and fourth rows, one can see the effect of origin separately for the two current residence possibilities. These differences are all consistent in direction and similar in size. The next to the last line of the table gives the average origin difference over the two current residence groupings, values ranging from 7 to 12 percentage points. The last line gives corresponding average differences between groups similar in origin but different in current residence (line 1 compared with 3, and 2 compared with 4). These values are also consistent, and about three times the size of the average origin differences. Thus current residence

Respondents Born in, Living in, and Preferring a Residence Type"

City over 50,000 Residence of parents when born Current residence Preferred residence

Other, Less Than 30 Miles from City

Other, More Than 30 Miles from City

30%

30% 44

35

40% 21

25

55

20

Total

100% 100 100

a- N • 1,380 respondents who answered these three questions. Sour.ce: National Opinion Research Center Amalgam Survey conducted in November, 1972.

499

Residential Preferences and Population Distribution

TABLE 6.-Percent of Respondents Preferring a Type of Residence by Current Residence and Residence of Birth, United States, 1972 (Value in Each Cell is Percent of Respondents Preferring This Residence Type)

Current Residence and Residence of Birth Current residence in area indicated Origin In area Not in area

Type of Residence Big City

Near

Away

57%

84%

74

49% 41

14

53

16

6

39

U

12 39

12

7 31

40

Current residence not in area indicated Origin In area Not in area Average differences Origin effect Current residence effect

33

Source: National Opinion Research Center Amalgam Survey conducted in November, 1972.

has an important effect in that people who live in an area are more likely to prefer it than people who do not, regardless of origin. But place of origin, representing an influence which may be more remote in time, also has a smaller but consistent effect on people's preferences for those preferring to live in a big city, near a big city, or farther away. Mazie and Rawlings (1972) reported similar results with their residence categories. REASONS FOR PREFERENCES

After stating their preference for a community size and location, respondents were read a list of reasons people might have for preferring to live in one kind of community or another and asked whether or not each was one of their reasons. The percent of people reporting each reason was tabulated by the three location types and graphed in Figure 1. In order to clarify the distinction between locations, the reasons were ordered by the percent of people who preferred to live away from a big city reporting them. These ranged from 13 percent for

higher wages to 89 percent for less crime and danger. This ordering gives a clear discrimination between big city and other preferences. Those preferring small towns and rural areas near big cities have almost the same pattern of response as persons preferring more remote locations. People who want to live near big cities, then, seek the same qualities as those preferring to live farther away; but both groups differ markedly from those who prefer large cities. This latter group differentially seeks higher wages or salaries, better job opportunities, the possibility of contacts with a variety of people, better schools, nearness to family or friends, and recreation or cultural facilities. Almost all those preferring more rural locations mentioned quality of life factors often associated with the country: less crime, quality air and water, better place for children. On the other hand, only about 20 percent favored more rural locations because of better job opportunities or higher wages there. These

500

DEMOGRAPHY, volume 12, number 3, August 1975

\~

LESS CRIME QUALITY AIR AND WATER

•• ••

.~

BETTER FOR CHILDREN

,.••• ." ......

....

LOWER COST OF LIVING RECREATION OR CULTURE

•••••

....

....... .•..•

I •.•••

NEAR FAMILY OR FRIENDS

I :

4J. \ .•.

BETTER SCHOOLS CONTACTS VARIETY PEOPLE

""

r

BETTER JOBS

1 -,,"

o

20

• .. ••

...

*,,"

1

HIGHER WAGES

.....

i

..

•••

, 40

60

80 100

PERCENT OF TOTAL

RESIDENCE PREFERRED •••••••• BIG CITY "'--NEAR • AWAY i.-Proportion of Respondents Stating a Reason is Important in Their Residential Preference by Type of Preference. Reasons Have Been Ranked According to Proportion of Mention by Respondents Who Prefer an "Away" Location

FIGURE

five reasons showed the maximum differences between the urban and rural preference groups.

The analysis of reasons was extended by controlling for present location. Present location made little difference, how-

501

Residential Preferences and Population Distribution

ever, in comparison to the differences between preferences for big cities and more rural areas. The people wanting to live in a different area thus respond in terms of the same qualities as those already there who wish to remain. Tables of reasons also were constructed separately by categories of age, income, and sex, but differences were small and not systematic. Perhaps people are simply responding to generally recognized stereotypes concerning the types of residences they prefer. The fact that over 40 percent of those preferring big cities gave reasons of less crime and danger there, and better quality of air and water, suggests that many respondents were simply assenting to factors desirable in any community. Nevertheless, systematic differences did emerge between those preferring big cities and those preferring other locations. A thorough analysis of possible reasons would require considerably more interview time than was available to us for detailed questions tailored to specific preferences.

somewhat higher proportion preferring remote rural areas (26 versus 20 percent) than the trichotomous classification of responses based on the two initial residential preference questions. Such response differences are to be expected in survey research, and the general conclusions of the study are the same with either preference measure. The pattern of first and second choices is of interest here. Whereas 43 percent of the respondents ranked a near location first and a more rural location second, only 12 percent ranked a near location first and a big city second (compare line 4 and line 3). In all only 32 percent ranked the big city either first or second, whereas 71 percent ranked an away location either first or second, and 97 percent ranked the near location first or second. Such results indicate that the desire for a small town or rural setting is extremely pervasive and that most respondents view the advantages of living in such a setting even at some distance from a large city as preferable to living in a large city.

THE RANKING OF PREFERENCES

An unanswered question is whether the basic preference pattern described in our data is for the metropolitan area itself (with the smaller city or open country therein viewed as a, more pleasant living alternative) ; or whether basic preference is really for small towns or rural areas themselves (with proximity to a big city representing a kind of cost paid for urban advantage). In an effort to ferret out this subtle distinction, we asked respondents to rank in order of preference three types of location: a city over 50,000, a smaller city or rural area within 30 miles of a city over 50,000, and a smaller city or rural area more than 30 miles from a city over 50,000. The results are given in Table 7 for persons classified by current residence. The firstrank choices to these questions show a somewhat lower proportion preferring large cities (19 versus 25 percent) and a

DISCUSSION AND SUMMARY

Antiurbanism in America appears to be qualified: although many people do not prefer to live in big cities, few want to live far from one. Previous survey data on residential preferences considerably overestimated the popularity of nonmetropolitan rural environments by failing to define them precisely in people's minds. When respondents are allowed to express a preference for the degree of proximity to a large city of over 50,000 population, they favor the peripheral metropolitan ring areas that have, in fact, been gaining rapidly in population. These findings have a direct meaning for policy, since they call into question arguments for population dispersal into nonmetropolitan areas based on public preferences alone. Other worthy objectives may be served by such a national

502 TABLE

DEMOGRAPHY, volume 12, number 3, August 1975

7.-Preference Ranking of Residence Types by Current Residence, United States, 1972 Preference Ranking by Respondent Big City Near Away First First Second Third Second Third

Second Third First First Third Second

Third Second Third Second First First

Total Number

Big City 33% 4 17 32 1 13 100 637

Current Residence Near Away 4% 1 9 60 1 25 100 500

3% 5 42 0 50 100 302

Total 17% 2 12 43 1 25 100 1,439

Source: National Opinion Research Center Amalgam Survey conducted in November, 1972.

goal, but the proportion of people eager to move to a remote nonmetropolitan setting appears to be small and balanced by an equal number already in nonmetropolitan areas who want to move closer to a big city. The findings illuminate an issue of population distribution policy which, like the controversy about whether or not voluntary family planning can suffice to achieve lowered growth rates, hinges on the matter of what voluntarism would produce. Our research suggests that, if people were to sort themselves out into the kinds of residential environments they claim to prefer, there would be no massive exodus to remote areas. Americans' residential preferences, when properly interpreted, lend no credence to the view that measures aiding dispersal to nonmetropolitan areas would simply be satisfying a large unmet public need, although such measures may be justified on other grounds. Despite the massive movements into metropolitan areas during this century, we found little evidence of a desire by migrants to return to their place of rural origin. We found that origins influence preferences but that present location is considerably more important. Furthermore, this small origin effect prevailed for those born in all three types of location considered, not just in rural areas.

As Mazie and Rawlings (1972) point out, the place-of-origin effect may move aggregate preferences toward big cities for later generations having higher proportions born there. These conclusions should not minimize the favorable orientation to rural and small town life expressed by the respondents on this as well as previous surveys. The reasons given here for preferences show that the commonly held advantages of rural life are believed to be present even within proximity of a large city. The ranking question showed that four out of five of the respondents who preferred to live near large cities as a first choice gave rural areas away from large cities as a second choice, rather than the large city itself. Among all respondents two-thirds ranked the large city least desirable of the three options. The discrepancy between preference and current residence which we found is consistent with trends in U. S. population redistribution over the past several decades. But as yet there is little direct evidence concerning the interrelations between residential preferences and migration behavior or between both of these and other attitudinal variables such as community satisfaction. Such knowledge would increase our understanding of population movements and is needed by those interested in formulating and at-

Residential Preferences and Population Distribution

tempting to implement population distribution policies on a national or local level. ACKNOWLEDGMENTS

This work was supported by the College of Agricultural and Life Sciences, University of Wisconsin, Madison, through a cooperative agreement with the Economic Development Division, U. S. Department of Agriculture; by the Graduate School Research Committee, University of Wisconsin; by the Agricultural Experiment Station (Article No. 6605), the College of Social Science, Michigan Sta.te University; and by the Center for the Study of Metropolitan Problems, NIMH (MH23489-01). Our questions were included on the November 1972 Amalgam Survey of the National Opinion Research Center, Chicago, Illinois. Analysis was done on the computer of the Center for Demography and Ecology, University of Wisconsin, provided through a grant from the Center for Population Research of the National Institute of Child Health and Human Development. We are grateful to Calvin L. Beale, David R. Brown, Philip Groth, Gordon De Jong, and especially two anonymous reviewers, who made helpful comments on earlier drafts of this paper. REFERENCES

Ad Hoc Subcommittee on Urban Growth. 1969. Population Trends. Hearings for use of the Committee on Banking and Currency, House of Representatives. Washington, D. C.: U. S. Government Printing Office. Advisory Commission on Intergovernmental Relations. 1968. Urban and Rural America: Policies for Future Growth. Report A-32. Washington, D. C.: U. S. Government Printing Office. American Institute of Public Opinion. 1970. Gallup Opinion Index, Report No. 57. Princeton: The Gallup Organization, Inc. Beale, Calvin L. 1972. Rural and Nonmetropolitan Population Trends of Significance to National Population Policy. Pp. 665-677 in Sara Mills Mazie (ed.) , Commission on

503

Population Growth and the American Future, Vol. V, Population, Distribution and Policy. Washington, D. C.: U. S. Government Printing Office. - - - . 1974. Rural Development: Population and Settlement Prospects. Journal of Soil and Water Conservation 29:23-27. Commission on Population Growth and the American Future. 1972. Population and the American Future. Washington, D. C.: U. S. Government Printing Office. Dillman, Don A., and Russell P. Dobash. 1972. Preferences for Community Living and Their Implications for Population Redistribution. Washington Agricultural Experiment Station, Bulletin 764. Pullman: Washington State University. Fuguitt, Glenn V. 1971. The Places Left Behind: Population Trends and Policy for Rural America. Rural Sociology 36:449-470. Hansen, Niles M. 1970. Rural Poverty and the Urban Crisis: A Strategy for Regional Development. Bloomington: Indiana University Press. - - . 1973. The Future of Nonmetropolitan America: Studies in the Reversal of Rural and Small Town Population Decline. Lexington: Lexington Books. Louis Harris and Associates, Inc. 1970. A Survey of Public Attitudes towards Urban Problems and toward the Impact of Scientific and Technical Developments. Study No. 2040. New York: Louis Harris and Associates, Inc. King, Benjamin F., and Carol Richards. 1972. The 1972 NORC Probability: Sample. Unpublished preliminary draft. Chicago: National Opinion Research Center. Lee, Everett S., J. C. Bresee, K. P. Nelson, and D. A. Patterson. 1971. An Introduction to Urban Decentralization Research. Department of Housing and Urban Development ORNL-HUD-3. Oak Ridge: Oak Ridge National Laboratory. Mazie, Sara Mills, and Steve Rawlings. 1972. Public Attitude towards Population Distribution Issues. Pp. 599--615 in Sara Mills Mazie (ed.) , Commission on Population Growth and the American Future, Vol. 5, Population, Distribution, and Policy. Washington, D. C.: U. S. Government Printing Office. Morrison, Peter A. 1970. The Rationale for a Policy on Population Distribution. Santa Monica: The RAND Corporation. National Area Development Institute. 1973. Public Opinion Favors Nonmetro Areas. Area Development Interchange 3:1, ff. National Goals Research Staff. 1970. Toward Balanced Growth: Quantity with Quality. Washington, D. C.: U. S. Government Printing Office. National Wildlife Federation. 1969. The U. S.

504

DEMOGRAPHY, volume 12, number 3, August 1975

Public Considers Its Environment. A National Opinion Trends Report. Princeton: The Gallup Organization, Inc. President's National Advisory Commission on Rural Poverty. 1967. People Left Behind. Washington, D. C.: U. S. Government Printing Office. U. S. Bureau of the Census. 1974. Mobility of the Population of the United States: March 1970 to March 1973. Current Population Re-

ports, Series P-20, No. 262. Washington, D. C.: U. S. Government Printing Office. Zuiches, James J., and Glenn V. Fuguitt. 1972. Residential Preferences: Implications for Population Redistribution in Nonmetropolitan Areas. Pp, 617-630 in Sara Mills Mazie (ed.). Commission on Population Growth and the American Future, Vol. V, Population, Distribution, and Policy. Washington, D. C.: U. S. Government Printing Office.