The Contribution of (ommon Land to Sense of Community
Transcription
The Contribution of (ommon Land to Sense of Community
The Contribution of (ommon Land to Sense of Community Ching Li College of Environmental Science and Forestry, State University of New York The purpose of this study is to determine the contributions of common land to sense of community in Taipei, Taiwan. The researcher defined four types of communities based on different types ofcommon land in communities. In each type, the researchers selected one typical community. In the Wan-Ho community (Type 1), most of the buildings are lower than five stories with community alleys, green space, and parking lots. In the Kao-Tai community (Type II), aU buildings are seven stories each with a building entrance, and an underground parking lots. In the Chen-Kung community (Type III) the buildings are from seven to 19 stories with courtyards, playgrounds, parking lots, and a community center. The Lon-Sam-Lin community (Type IV) is a single family housing community with a main secured entrance, a golfpractice green, a swimming pool, playgrounds, tennis courts, basketbail courts, and a community center. The researcher conducted the survey with the lB-item scale modified from the Nasar and Julian (1995) scale to measure sense of community and also study the participation frequency of activities on the common land. There were 529 respondents (Type 1=106, Type II =102, TYpe III =205, Type IV=115). The overall sense of community among these four communities was different. The respondents from the Type III community have more sense of membership, integration and fulfillment needs, and sharing emotional connection. The respondents from the Type II have alower score on sense of community. We also focused on how the activities on the common lands influenced sense of community. The researchers found that the daily activities play an important role to influence sense of community, if these activities also create other interactions. We suggest further studies to observe the space outlets of the activities in order to determine what types ofdesign ideas will contribute to sense of community. Introduction Common land provides spaces for formal or informal meetings, recreational uses, or aesthetic appreciation in a community (Nelessen, 1992). It is fully aware that these places may serve different age groups in the community. In 1977, Alexander pointed out "Without common land no social system can survive:' In Taiwan, based on the Building Complex Management Act (RCM) of 1995, a community with common land must set up a management committee. According to this legal approach, common land can help local people to clearly define their community boundary and to applypossible resources to develop their community. On the other hand, the common land in the communities becomes different to other public spaces. The community common land exclusively belongs to the community members. AIso, each member in the corn munity has his or her rights to access the corn mon land within a desired manner. However, this desired manner is shaped by culture, competing interests, and regulations (Carr et al., 1992). The uses of the common land may reflect the agreement among members in their community. According to different community development concepts in Taiwan, corn mon land has a variety forms and functions in communities. Therefore, the study attempted to determine if people living in communities with different types of common land feel differently regarding sense of community. Study Area The researcher studied four communities in Taipei, Taiwan area. These four communities are: l)The Type 1 Community (Wan-ho Community): There are 1288 households in this community. This community is a typical early corn munit y design (Figure 1). refereed papers . 57 The major parts of this community include the military family housings and the Bureau of Forest family housings. The community is next to a river, but beyond the river is a city highway and there is about a ten-meter-high wall between the river and the community edge. The community uses the space under the highway to create parking space, green belts and sorne small plazas. Most of the buildings in this community are lower than five stories. ln the military housings, each house entrance faces to each other, and the average width of alleys is about Figure 1 Wan"Ho Community Appearance and Site Plan walked at the corn munit y alley, it is verv people talking in their living room and people were doing in there. Therefore, observation, sorne residents went out and me why 1 was there. These houses sur round a small square and the management committee office. ln the Bureau of Forest family housings, it is a fourstory building complex. Each apartment is about 1000 to 1500 square feet. The building complex also has a small plaza. Right now, the plaza is occupied bya food vender and becomes a community parking lot. Other buildings were built by other unknown builders around 1970. They very flexibly use and manage their common space. In military housing are a, it is very hard for cars to park or run into sorne community alleys. The alleys become a playground for children, a place for neighbors to get together, and extra storage space. Additionally, in the green space, members put their old sofa sets for getting together in the summertime. During the special event, they clean up the green space and arrange sorne temporary facilities to hold a party. They also hold community meetings in the green space. 58 . edra 29/1998 Figure 2 Kao-Tai Community Appearance and Site Plan 2)The Type II Community (Kao-Tai Community): There are 637 households in this eommunity. The apartment size is about 1400 square feet to 2000 square feet. People live here beeause of their ehoiees. The whole eommunity was built at the same time and aH buildings are seven stories in straight, paralleI rows (Figure 2). In eaeh building, approximately 14 households share a public entrance and an elevator. The whole eommunity shares an underground parking lot. The eommunityalleys are about eight meters and beeome extra parking space for this area. Some of residents modify their front yards to beeome their parking space. There is no sidewalk for pedestrian in eommunity alleys. The pedestrian needs to wateh traffie when walking around the community. \ refereed papers . 59 Figure 3 Chen-Kung Community Appearance and Site Plan Communhy 3) The Type III Community (Chen-Kung Community): This corn munit y is a typical super block development. Traffie system is rigidly spared. Almost aH of automobiles need to park outside the community or in community parking lots. Therefore, people can walk around the communitywithout any interceptions. Three major building complexes are from seven to 19-stories high (Figure 3). These building complexes sur round the center court yards, and each building complex has its own yard and public entrances. In this community, haHways and green belts are used to connect each building. In sorne buildings, the residents hire door attendants from 8:00 a.m. to 5:00 p.m. The whole community shares a community center, and playgrounds. About 2452 households are in this community. The sizes of apartments are about 900 to 1600 square feet. court yards .J....l..._ _ _....J ~~=~~;~~~~-~~ '-- ~ la n rilt __StOl, o s.vent_Sl«y iil Hillet_Slor, I:.'l N!Ii: flClîl)' 0.- 60 . edra 29/1998 s.a- ::igure 4 Lon-Sam-Lin Community Appearance and Site Plan 4) The Type IV Community (Lon-Sam-Lin Community ): This is a new style of community. There are 1157 households in this community. All the buildings in the community are single family housings (Figure 4). Each house has its own entrance and yard. The range of the living area in a house is from 2845 square feet to 7576 square feet. The automobile and pedestrian routes are separated. There is a main road connecting each community alley. Each community alley has agate to stop cars into the alleys and lead aU cars into an underground parking lot entrance. Pedestrian routes include sidewalks on the main road and trails connecting each common land in the community. The shuttle buses go around the community or commute to other placep. The community has a main secured entrance, a golf practice green, a swimming pool, playgrounds, tennis courts, basket baIl courts, and a corn munit y center. refereed papers ·61 Method Pretest results First, the researcher did the observations and interviews in these four communities to understand the activities on community common land. The measurement of sense of community was developed through modifying the Nasar and Julian's measurement in 1995 and sorne other related research, such as Glynn's study in 1985, and Davidson and Cotter's studies in 1993(Table 2). Then, the researcher did the pretest for the measurement of sense of community. The pretest discussed the reliability and validity of the scale for measuring sense of community. After the pretest, the researcher conducted a survey that includes the measurement of sense of community, participation frequency of activities on common land, and demographic information in these four communities. The researchers randomly mailed the survey through the list of community members. Then, the researchers waited for two weeks and called each selected subject to remind him/her to return the questionnaire. Results trom Observation on Community Activities During the observation process, the researcher found the daily activities may create other social interactions. Sorne daily activities create a lot of social interactions, such as talking to each other, and playing together. However, sorne do not create many social interactions, such as looking and greeting. In the community, the management committee arranges sorne community clubs to provide opportunities for members to meet others who have the same interests. During the special time, the management committee may hold celebrations. Additionally, the members participate in the management processes, so sorne activities, related to the management process, are labeled as the management activities. Therefore, the researcher coded the activities into five groups (Table 1): • Daily Activities - Low Social Interactions (DAL) • Daily Activities - High Social Interaction (DAH) • Special Activities (SP) • Rotational Activities (ROT) • Management Activities (MANA) 62 . edra 29/1998 Although the previous researchers tested the reliability and validity of the scale in the United States, this study still pretested for this measurement scale because of the culture difference. There are three processes in this pretest: 1) Translation Test: First, the researcher translated these 18 questions from English to Chinese. Then, the researchers found five graduate students from the telephone list of Chinese Students Association from Taiwan in the Syracuse area. The researcher asked these five Chinese students translated these questions from Chine se to English. After this process, the researcher found these translated questions can precisely match the original English questions. This indicates the questions in Chine se and in English are similar in meaningto Chinese respondents. 2) Reliability Test: The researcher did the reliability test for this measurement. The pretest with 100 respondents holds in two small communities. One is similar to the Type II. The other one is similar to Type III. The Cronbach Coefficient Alpha estimate of reliabilityfor the l8-item scale equal to 0.78 that is larger than 0.70 that was recommended as the minimum value by Nunnally (Hatcher, 1994). However, the Cronbach Coefficient Alpha is 0.53 for estimating the reliability of the Il-item scale. This means respondents can respond to the 18-item scale more consistently than the Il-item scale. If S3 and S6 were dropping, Cronbach Alpha would become higher (Table 3). However, the researcher found these two questions are reversed questions, so the respondents cannot respond consistently. Because the l8-item scale has reached the requirement for reliability, the researcher would like to retain these two questions to do further study. 3) Validity Test: Referring to Babbie's "Survey Research Methods" in 1990, the validity consists of face validity, criterion-related validity, content validity, and construct validity. The face validity indicates the Table 1 Activities on Common land Daily Activities -low Social Interaction (DAll: Type 1 Type Il Type III Type IV 1meet my neighbors at my community parking lots. Then, 1just greet them and leave ..................... 1 .............. 1 ............ 1 ............. 1 1meet my neighbors at the hallways. Then, 1just greet them and leave ............................................ 3 .............. 3 ............ 3 Wh en 1pick up my mail, 1meet my neighbors and greet them .......................................................... 7 .............. 7 ............ 7 ............. 7 1park my car in the community parking lots ...................................................................................... 9.............. 9 ............ 9 1put my personal belongings at the hallway ................................................................................... 10 ............ 10 .......... 10 ........... 10 1talk awalk at the community alleys ..........................................................................................:.... 11 ............ 11 .......... 11 ........... 11 1do exercises at alleys ...................................................................................................................... 13 ............ 12 .......... 12 ........... 12 1park my car at the community alleys ............................................................................................. 17 ............................................... 9 1do exercise in the green spa ce ....................................................................................................... 23 1take awalk at the community green space .................................................................................... 24 ln the elevator, 1just greet my neighbors but not talk to each other .................................................................... 5 ............ 5 1do exercises at the community courtyard ........................................................................................................................ 14 ........... 14 1go swimming at the community swimming pool ............................................................................................................................... 3 1play tennis at the community tennis court ......................................................................................................................................... 5 1take the community bus ................................................................................................................................................................... 24 1do exercises at the community basketball courts .............................................................................................................................. 37 1do exercise at the community skate rink ........................................................................................................................................... 40 Daily Activities - High Social Connect (DAH): Type 1 Type Il Type III Type IV 1meet my neighbors at my community parking lots. Then, 1talk to them for awhile ........................ 2 .............. 2 ............ 2 ............. 2 1chat with my neighbors for awhile at the hallway .......................................................................... 4 .............. 4 ............ 4 When 1pick up my mail, 1meet my neighbors and talk to them for awhile ....................................... 8 .............. 8 ............ 8 ............. 8 1buy something and chat with my neighbors at the community alleys. Children in my family play with others at the community alleys ...................................................... 14 ............ 13 .......... 13 ........... 13 1encourage children in myfamily to play with neighbors' children. Also,1 connect with their families' members ................................................................................ 15 ............ 14 .......... 17 ........... 17 1chat with my neighbors at the community green space ................................................................ 22 Children in my family play at the community green spa ce ............................................................... 25 ln the elevator, 1chat with my neighbors ............................................................................................................. 6 ............ 6 Children in my family play at the community playground ................................................................................................. 15 ........... 15 1chat with my neighbors at the community playground ................................................................................................... 16 ........... 16 1chat with my neighbors at the courtyard ......................................................................................................................... 36 1chat with my neighbors at the community swimming pool ..............................,.............................. 4 1chat with my neighbors at the community tennis court .................................................................. 6 1chat with my neighbors at the community bus .............................................................................. 25 Children in my family play at the community basketball courts ....................................................... 38 1chat with my neighbors at the community basketball courts ........................................................ 39 Children in myfamily play at the community skate rink .................................................................. 41 1chat with my neighbors at the community skate rink .................................................................... 42 *the number in the cell is the question number in the questionnaire refereed pa pers . 63 Table 1 Activities on Corn mon land (continued) Special Activities (SP): Type 1 Type Il Type III Type IV 1participate special events on the green space .................................................................................. 6 1hold my personal events, such as parties and special ceremonies, at alleys .................................... 28 ............................................. pl 1 participate special events on the community center ........................................................................................................ 38 ........... 35 1participate the recreational sport tourna ment in my community ..................................................................................................... 34 1participate the community annually hiking ........... ;.......................................................................................................................... 36 **p1:This activity is listed as aprohibited activity in the community management principles. Rotational Activities (ROT): Type 1 Type Il Type III Type IV participate the community meeting .............................................................................................. 19 ............ 19 .......... 22 ........... 22 participate the party to worship the Earth God each month .......................................................... 21 participate the drawing club at community center ........................................................................ 26 ............................ 30 ........... 30 participate the dancing club at the community center .................................................................. 27 ............................ 31 participate the English conversation group at the community center .............................................................................. 24 ........... 27 participate Mom's singing club at the corn munity center ................................................................................................ 25 ........... 26 participate the Taiwanese singing club at the community center .................................................................................... 26 participate the Chinese opera club at the community center ........................................................................................... 27 participate the 60s' song club at the community center ................................................................................................... 28 participate the Chinese craft club at the community center ............................................................................................. 29 ........... 32 participate the flower-arranging club at the community center ...................................................................................... 33 participate the Chinese folk music club at the community center .................................................................................... 32 participate the chess club at the community center ......................................................................................................... 34 participate the reading club at the community center ..................................................................................................... 35 ........... 31 participate the cooking club at the community center ...................................................................................................................... 28 participate the Japanese conversation group at the community center ............................................................................................ 29 Management Activities (MANA): Type 1 Type Il Type III Type IV 1participate the planning processes for the special events in my community .................................... 5 ............ 18 ......... 37 ............ 33 1help fix sorne problems in my community, if 1know ...................................................................... 16 ............ 15 ......... 18 ............ 18 1participate management affairs .................................................................................................... 18 ........................... 21 ............ 21 1suggest sorne ideas to change my community .............................................................................. 20 ............ 20 ......... 23 ............ 23 1am not willing to pay for the management fee ................................................................................................ 17 ......... 20 ............ 20 1help the management committee ................................................................................................................... 16 ......... 19 ............ 19 *The number in the cell is the question number in the questionnaire 64 . edra 29/1998 Table 2 The Measurement of 5ense and Community The Questions: 51 52 53* 54 55 56* 57 58 59 510 511 512 513 514 515 516 517 1am quite similar to most people who live here .................................................................................................................... NJ ...... ( If 1feellike talking, 1(an easily find someone in this corn munity to talk to ........................................................................... NJ ...... ( 1DO NOT (are whether this (ommunity does weil or not ....................................................................................................... NJ ...... ( 1am willing to help other people who live in this community if they are in trouble .............................................................. NJ ...... c 50me of neighbors in this community are my friends ...................................................................................................................... c If 1am upset about something personal, there is NO ONE in this (omm unity whom 1can turn to .......................................... NJ ...... c 1have sorne friends in this community whom 1can rely on ................................................................................................... NJ ...... c If there were serious problems in this corn munity, the people in this community could get together and solve them .......... NJ ...... c If 1had an emergency, even people 1do not know in this corn munity would be willing to help ............................................ NJ ...... ( If someone does something good for the community, that makes me feel good ................................................................... NJ ...... c 1feel my value is similar to the rest of the people in the community ................................................................................................ ( My (oncerns can get quick responses in this community .................................................................................................................. c 1feell belong here ........................................................................................................................................................................... c 1am proud totell otherwhere 1live ................................................................................................................................................. c The community can satisfy what 1need in my daily life ................................................................................................................... c It will take a lot from me if 1 move out this community .................................................................................................................... ( 1feel safe in this community ............................................................................................................................................................ ( 518 1can tell who do es or doesn't live in the community ........................................................................................................................ ( The police in this neighborhood are generally friendly .......................................................................................................... NJ NJ: the questions are developed by Nasar and Julian (1995) ............................................................................................................................................................... c the questions are used in this study common agreement on this measurement. This scale has been used or tested in many studies, such as in Nasar and Julian's studies in 1995, Glynn's studies in 1981 and Davidson and Cotter's studies in 1993. Therefore, the researcher believes the scale has face validity. The criterion-related validity means the predictive validity. The researcher followed the McMillan and Chavis's four elements of sense of community to structure the scale. The McMiIlan and Chavis's theory about sense of community has been widely used in sense of community studies. In the scale, each element is measured by three to five items (Table 4). The scale should have criterion-related validity to measure sense of community. AIso, from the correlation analysis, the range of correlation coefficients is from -0.00304 (s5-s18) to 0.66387 (sl-s11). It indicates the relationships in each pair of variables are low to moderately related. It shows most items have their own directions to measuresense of community and also correlate to the total measurement (Table 3). The content validity means the studies also need to coyer the other related information in this study. In this scale, the researcher only measured sense of community, but in the questionnaire, the researcher also coUected demographic information, and participation frequency of corn munit y activities. The researcher thinks the whole studyshould have content validity. In order to prove the construct validity of the scale, the researcher applied MANOVA (Multivariate Analysis of Variance ) to test the ability of the scale to differentiate the sense of community between these two communities. The results show the 18item scale can differentiate these two communities (F=4.0081 and the Pr < 0.05). refereed papers . 65 Table 3 Results from Pretest: The Cronbach Coefficient Alpha Raw Variables 1 5tandardized Variables 2 Table 4 The Theoretical Elements in 5ense of Community The overall sense of community (18-item scale): Cronbach Coefficient Alpha 0.855114 for four communities survey. Oeleted Variable Correlation withTotal Alpha Correlation with Total Alpha 51 0.421185 0.753444 0.442794 0.763129 52 0.494484 0.745841 0.489077 0.759735 53 0.029542 0.783625 0.059841 0.789816 54 0.366664 0.757872 0.3847·18 0.767335 55 0.178107 0.770858 0.184963 0.781365 56 -0.413991 0.808639 -0.375719 0.817296 57 0.397905 0.754960 0.395874 0.766532 514 1am proud to tell others where 1live. 58 0.390200 0.755721 0.362307 0.768943 517 1feel safe in this community. 59 0.346202 0.760212 0.353333 0.769584 518 1can tell who does or doesn't live in the community. 510 0.367386 0.758275 0.386925 0.767177 511 0.659213 0.738176 0.675995 0.745647 512 0.625336 0.737957 0.631023 0.749093 513 0.593724 0.742704 0.610915 0.750622 514 0.501260 0.753525 0.514575 0.757850 515 0.244767 0.768882 0.211606 0.779533 516 0.370415 0.757283 0.343324 0.770298 517 0.455752 0.749750 0.426886 0.764287 510 If someone does something good for the community, that makes me feel good. 518 0.441086 0.751472 0.422387 0.764614 512 My concerns can get quick responses in this community. 1 for raw variables: 0.769556 lfor standardized variables: 0.779425 In short, this 18-item scale has the reliabilityand validity to measure sense of community in Taiwan. Therefore, the researcher has confidence to applythe scale to further studies in the Taipei area. Membership (5M): 51 1am quite similar to most people who live here. 511 1feel my value is similar to the rest of people in the community. 513 1feell belong here. 2 Influence (51): 53 1DO NOT care whether this corn munity does weil or not. 58 If there were serious problems in this community, the people could gettogether and solve them. 3 Integration and fulfillment of needs (5IF): 57 1have sorne friends in this corn munity whom 1can rely on. 59 If 1had an emergency, even people 1do not know in this corn munitywould be willing to help. 515 The community satisfies what 1need in my daily life. 4 Share Emotional 5haring (55): Results from the Community Survey The study conducted a survey in these four communities. The survey includes the 18-item scale to measure sense of community, questions to study the participation frequency of activities on the common land, and questions about the demographic data. The researcher sent 2500 questionnaires to the selected communities (Type 1 =500, Type II = 500, Type III = 1000, and Type IV = 500). The total respondents to the questionnaire was 598, but the to- 66 . edra 29/1998 52 If 1feellike talking,l can generally find someone in this community to talk to. 54 1am willing to help other people who live in this community if they are in trouble. 55 50me of neighbors in this community are my friends. 56 If 1am upset about something persona!, there is NO ONE in this community whom 1can turn. 516 It will take a lot from me if 1move out this community. tal of valid respondents was 528 (Type 1= 106, Type II = 102, Type II = 205 Type IV = 115). The return rate of this survey is 21.12%. From the mode (Table 5), we can find the respondents from the Type 1 community have a lower education level, are more liberal in political orientation, and have been living in the community longer. In the Type II community, respondents work for service industry. In the Type III community, the respondents are similar to the whole respondents. The Type IV community respondents have higher income and most of them are business-persons and students. However, there were 140 respondents who refused to answer this question. In addition, most respondents are housewives, so they daim they do not have any income. Almost 40% of respondents daim they do not have income or refuse to answer the questions. Although the statistic shows the income level is associated to the types of communities (df = 15 F value =37.37 pr<0.05), the researcher thinks that it may not be an appropriate measurement. From Chi-square test, the types of communities are associated to many demographic variables. These characteristics are age, education level, political orientation, the length of residence, occupation, the number of generations living together, marital status, and ownership. It implies that each type of communit y has its own demographic characteristics. Table 5 The Demographie Characteristies* of Respondents Among These Four Communities (Mode) Age Education Level Politieal Orientation Length of Residence Occupations Generations Marital Status Ownership Income Level Type 1 Type Il Type III Type IV Overall 40-65 High school Liberal 20 years Housewives Two Married Own 10,000-20,000 40-65 University Moderate 12-14 years Service Two Married Own 10,000-20,000 40-65 University Moderate 10-12 years Housewives Two Married Own 10,000-20,000 40-65 University Moderate 4-6 years Students & Businessman Two Married Own 20,000-30,000 40-65 University Moderate 10-12 years Housewives Two Married Own 10,000-20,000 *all demographic variables are associated with types of communities p<0.05 Table 6 The Means of Sense of Community Y1* - Indieates Average Score of Sense of Community SM* - Indicates Sense of Membership SI** - Indieates Sense of Influences SIF* - Indieates Sense of Integration &Fulfillment of Needs SS* - Indicates Sense of Emotional Sharing *pr < 0.05 Type 1 Type Il Type III Type IV 3.72 3.60 3.55 3.36 3.86 3.73 3.63 3.68 3.58 df=54 3.92 3.54 3.70 3.71 3.92 3.50 3.62 df=3 df=3 df=3 3.98 3.62 3.70 3.55 df=3 f=4.06 f = 6.41 f=2.40 f = 3.72 f=2.76 **pr< 0.10 Bolded is used to indicate the highest score among these four communities. Italie is used to indicate the lowest score among these four communities. refereed pa pers . 67 Sense of Community Among Four Communities The purpose of this study is to test the effects of community types on sense of community. The researcher applied the Multivariate Analysis of Variance (MANOVA) to test the study hypothesis. In order to further analysis, the scores of sense of community are defined as: Strongly Agree =5, Agree =4, Not Sure ~3 Disagree =2 and Strongly Disagree = 1. Therefore, the higher score stands for the higher sense of community. The researcher applied the MANOVA to test the study hypothesis: • HO: In different types of communitiy, the means of sense of community (SI-S18) is equal. • HI: In different types of corn munit y, the means of sense of community (SI-SI8) are not equal. The results from the MANOVA show among these four communities the means of sense of communit y (sl-s18) are significantly different at 5% significant level (df =54 F value is 3.84 pr<0.05). It indicates these four communities have different means of sense of community. In addition, the researcher also tested the differences in the scores of average sense of community, and scores of the McMillan and Chavis's four elements of sense of community (Table 6). The results also show the me ans of average scores are different among these four communities. From the results, the respondents in the Type III have the highest overall sense of community. The respondents in the Type II have the lowest sense of community. The means of sense of membership score are also significantly different among these four communities at the 0.05 significant level (df =3 f=6.41). The mean scores of sense of integration and fulfùlment of needs are significantly different among these four communities at the 0.05 significant level (df =3 f =3.72). AIso, the me ans of sense of sharing emotional connections are different at the 0.05 significant level (df=3 f=2.76). From the results, the respondents from the Type III community have more sense of membership, inte- 68 . edra 29/1998 gration and fulfillment needs, and sharing or emotional connection. The respondents from the Type II feelless about three of the four elements of sense of community. There is no significant difference of sense of influence arnong these four communities at the 5% significant level. If we change the 5% significant level to the 10% of significant level, the means of sense of influence among these four communities are different. The respondents from Type 1 feel more sense of influence. Contributions of Activities on Common land to Sense of Community People more often use common land for their necessary activities, such as passing, parking, and picking up mail (Table 7). These simple daily activities also create sorne social interactions such as greeting, talking, and playing. These combined activities significantly influence the sense of community. The results from the correlation analysis show the relationship between sense of community and the participation frequency of daily activities with high social interactions has a higher correlation coefficient and also meets the 5% of significant level (Table 8). Sense of membership, sense of interaction and fulfillment of needs, and sense of emotional sharing also have higher correlation coefficient with daily activities with high social interactions. Only, sense of influence has higher correlation coefficient with management activities. The results infer that higher participation frequency of the daily activities with high social interactions may increase sense of community. People more frequentlyparticipate in common management pro cess, and feel they are able to influence the community. Among these four communities, respondents also have a different frequency of participation in activities on common land (Table 7). Respondents in Type III community have a higher frequency of participation daily activities with social interactions and in special activities, than in other communities. In the Type I community, respondents have a higher frequency of participation in management activi- lable 7 Sense of Community and Frequency of Activities Type 1 Type Il of Community Type III Type IV 3.72 3.55 3.73* 3.68 S<.?nse of Influence 3.98* 3.86 3.92 3.92 :>ense of Integration and Fulfillment 3.62 3.54 3.70* 3.50 :>ense of Sharing 3.70 3.55 3.71* 3.62 Daily Activities with Low Interactions 3.00 2.80 3.23* 2.99 Daily Activities with High 1nteractions 2.89 2.81 2.95* 2.74 Special Activities 2.54 0.00 2.85* 2.42 Rotational Activities 2.17* 0.00 1.41 1.45 Management Activities 2.87* 2.63 2.5 2.53 • indicates the highest score among these four communities Table 8 Correlation Between Sense of Community and the Freq uency of Activities ties and rotational activities. In the Type II communit y, they do not have special activities and rotational activities in their community. Discussion From ab ove results, we know the participation frequency of activities also reflects the sense of communit y. We found respondents in the super block community (Type III) have more sense of community than others. Additionally, people in the Type III communityalso participate more often in the daily activities with other social interactions (Table 7). The researcher implies that in the Type III community physical setting in the Type III community design, people need to walk into the community. Also, the pedestrian feels free to walk around the community and there are many benches on community courtyards, playgrounds or greenbelts. The residents also put sorne chairs or sofas in the hallways or around public entrance areas. People walk around the communitywithout consideration of traffic and also have facilities to take a break. These conditions may encourage people to interact with others, so in the Type III community, people feel more sense of community. Y1 SM SI SIF SS DAl 0.45337 0.0001 0.37196 0.0001 0.31593 0.0001 0.39549 0.0001 0.40463 0.0001 DAH 0.49074 0.0001 0.39883 0.0001 0.31565 0.0001 0.40242 0.0001 0.47862 0.0001 SP 0.31183 0.0001 0.29810 0.0001 0.22078 0.0001 0.21566 0.0001 0.26144 0.0001 People involving common land management process may also effect sense of influence. In the Type l community, respondents more often participate in rotational activities and management activities (Table 7). Additionally, the management committee was set up byvolunteer residents. Therefore, they have a higher score in sense of influence. ROT 0.18974 0.0001 0.18932 0.0001 0.13995 0.0013 0.11884 0.0063 0.15344 0.0004 Conclusion MANA 0.38916 0.0001 0.30109 0.0001 0.38766 0.0001 0.29140 0.0001 0.31483 0.0001 Yl indicates average score of sense of community 5M indicates sense of membership 51 indicates sense of influences 51F indicates sense ofintegration &fulfillment of needs 55 indicates sense of emotional sharing DAl indicates daily activities -Iow social interaction DAH indicates daily activities - high social interaction 5P indicates special activities ROT indicates rotational activities MANA indicates management activities Daily activities with other social interactions are important to inspiring sense of community. Involving common land management pro cess also contributes to sense of community. The physical setting also affects opportunities for these activities. Therefore, the researcher infers that if the corn munit y designers can create space, not only for necessary activities, but also for other social interactions, such as chatting and playing on common land, the sense of community would be stimulated. In addition, providing a space on common land, in which the com- refereed pa pers . 69 munit y members can get together and have chances to participate in the management process, may increase sense of community. Further research may study the space outlets of these activities to determine what types of design ide as currently affect sense of community. The research can help planners and designers to understand which kinds of conditions the activities can be evinced and how to apply the physical designs to increase sense of community. References Alexander, C. (1979) A Pattern Language. Oxford: Oxford University Press. Babbie, E. (1990) Survey Research Method. Belmont, CA: Wadsworth. Inc. Copper Marcus, C. (1965) Some Social Implications of House and Site Plan Design at Easter Hill Village: A Case Study. Berkeley, CA: University of California. Carr, S., Francis, M., Rivlin, 1. and Stone,A. (1992) Public Space. Cambridge: University press. Hatcher,1. (1996) A Step byStepApproach to Use the SAS Sys- tem for Factor Analysis and Structural Equation ModeZ. Carry, NC: SAS Institute Inc. Housing Dept. of Taipei City Government (1987) Housing Taipei from 1976 to 1985. Taipei, Taiwan: Taipei City Government. Lieberg, M. (1995) Teenagers and Public Space. Communication Research. 22(6).720-744. Nasar,J. 1. and Julian, D.A. (1995) The Psychological Sense of community in the Neighborhood. Journal of the American Planning Association. 61 (1).178-184. Nelessen,A.C. (1994) Visions for a New American Dream. Chicago, IL: Planner Press. 70 . edra 29/1998 The Impact of the Sociophysical Environment on Interpersonal Communication and Feelings of Belonging in Work Groups Bonnie N. Pollack Department of Psychology, University of North Carolina at Charlotte This study examined how properties of the physical environment and social climate influence group members' feelings of belonging as mediated by interperson al communication. Group-level properties of the physical environment included worker density and degree of work setting openness whereas group-level social climate properties were group cohesion and supervisor support. Mediating properties of communication networks were intensity of ties and enjoyment of communication among workers. Data were gathered from aU employees in nineteen non-customer service groups within four organizations using selfadministered questionnaires. Group sizes ranged from three to fourteen members, resulting in a total individual-level sample of 105 employees. Physical measurements of the work settings were also made. A hypothesized cross-level model was tested using hierarchicallinear modeling and path analysis. Of the four group-level predictors, only group cohesion had a significant positive direct effect on individuals' feelings of belonging. Intensity of ties and enjoyment of communication mediated the relationship between group cohesion and feelings of belonging. The two communication properties did not mediate the other three direct relationships. The hypothetical model was only partially supported. Furthermore, social climate had a greater impact on interpersonal communication and feelings of belonging than the physical environment. Participants In order to understand contextual influences of individual behavior and attitudes in the workplace, we need to examine both traditionally studied social environments as well as the physical setting. Few rese archers have studied how physical and social environments together affect interpersonal interactions and individual outcomes in work settings, despite the fact that Moos (1976) and Walsh (1987) note that the se different types of factors are likely to influence each other reciprocaUy and must be studied together. This study examines the relationships between social climate and physical environment antecedents and feelings of belonging as mediated by interpersonal communication in the context of work groups. The present studyextends previous research by: 1) examining both physical environment antecedents and social environment (climate) antecedents within the same study so that the relative importance of antecedents can be assessed; 2) examining the relationships between group-level phenomena and individual-level phenomena for individuals nested within groups; and 3) examining the outcome of employees' feelings of belonging within a work group as influenced by physical and social settings. One of many possible psychological reactions to the workplace environment (both social and physical) is workers' feelings ofbelonging. The contextual influences on employee feelings ofbelonging is an area in which further understanding is needed. Recently, Stokols (1990, pp. 644-645) has stated, ''An additional direction for future research is to ... identify the physical and social attributes of settings that contribute to individuals' experiences of spiritual enrichment. More specificaUy, what environmental and social arrangements are most closely associated with feelings of... social acceptance, and belongingness ....?" In the organizational behavior literature, belonging and affIliation are often presented in theories of employee motivation (e.g., Gibson, Ivancevich, & DonneUy, 1994). In classical psychological theories of motivation (Alderfer, 1972; Fromm, 1968; Herzberg, Mausner, & Synderman, 1959; Maslow, 1954; McClellend, 1971), belonging, or belongingness, has refereed papers . 71 been defined as a human need. Sociobiological and socioanalytic theories further suggest that humans have an innate need for belonging. For example, in Baumeister and Tice's (1990) exclusion theory, the underlying mechanism is the biologically based need to belong, which motivates people to be together with others. Based on the assumption that to sorne degree, people do have a basic need to belong, the present study focuses on feelings of belonging in the workplace as a consequence ot:environmental factors. Here, belonging represents a feeling of whether the need is met, versus the motivational need itself. Belonging, as experienced bywork group members, may be defined as a subjective feeling of being an integral and indispensable part of a work group, which is adapted from definitions by Anant (1970) and de Jong-Gierveld and Kamphuis (1985). sonal interaction and individual needs. The second aim of this research is to understand how the social climate factors of peer cohesion and staff support in conjunction with the physical environment properties affect interpersonal communication and individual feelings ofbelonging. The purpose of this research is to combine two streams of research on social climate antecedents with physical environment antecedents, and to understand how they, in combination, affect feelings ofbelonging in the workplace as mediated by interpersonal communication. The conceptual model combines the theoretical frameworks of Becker (1981) and Moos (1976). Becker's research focuses on the role of physical settings in organizations. He postulated a model in which sorne physical environment properties directly support individu al outcornes and others act as catalysts. Physical features that could act as catalysts include size, arrangement, quality, location, and density of settings. He proposes that these features indirectly affect individual outcomes as mediated by social interaction and communication (both the quality and level). Becker's model has not been empirically tested, and thus, one aim of this research is to test his proposition that physical environmental factors can act as catalysts. The conceptual model is shown in Figure 1. The model is mediational, as suggested by Becker's (1981) environment -as-catalyst process, with the interpersonal communication constructs mediating the relationships between the social and physical environmental constructs and feelings of belonging. Two group-Ievel physical environmental factors are included. They are density, which is the number of people per a given unit of are a (Becker, 1981), and openness, which is the ratio of the total square footage to the totallength of its interior walls (Oldham & Rotchford, 1983). Two group-Ievel social climate factors are included. They are group cohesion, which is the perception of togetherness or sharing among members within the work group setting (Koys & DeCotiis, 1991) and workers' tendency to stick together and help each other (Moos, 1976), and supervisor support, which is the perception of the tolerance of member behavior by superiors (Koys & DeCotiis, 1991). The two mediational communication constructs are intensity of ties, which is degree to which an individual has face-to-face communication with other members in a communication network (e.g., work group), and enjoyment of face- toface communication, which assesses how enjoyable face-to-face interactions with other work group members are, on average, for individuals (Monge, Edwards, & Kirste, 1983). The individu al hypotheses are indicated with numbers on the arrows, and predictions for the directions of the relationships are indicated with plus and minus signs. The other theoretical influence is Moos' (1976) conception of the social environment in work settings. Moos' social climate framework incorporates relationship dimensions, which identify the nature and intensity of personal relationships within the environment, as weIl as dimensions of personal development and system maintenance. Moos postulated that the social relationship dimensions of involvement, peer cohesion, and staff support in the workplace are the dimensions that most affect interper- The conceptual model proposed is an attempt to further the theoretical understanding of the multilevel phenomena of interest. Until recently, there has been little theory-driven research on multi-Ievel phenomenon (Rousseau, 1985). It is important to develop and test theory at both levels of analysis and not just focus on group-Ievel or individual-Ievel phenomena alone. Employees' behaviors and attitudes are partially determined by the groups within which they work, and as such, those contextual ef- 72 . edra 29/1998 fects should be examined. The current conceptual model is a cross-level model (Klein, Dansereau, & Hall, 1994; Rousseau, 1985), which is a popular subtype derived from the traditions of ecological psychology(Barker, 1968; Wicker, 1992),in which contextual characteristics affect behavior. Method Participants Participants were employees working at least halftime (i.e., 20 hours per week) who were members of work groups as defined below. Employees who were members of more than one work group were included if they identified the selected group as their group of primary affiliation. Once groups were selected by the method described below, aU employees, including the group's supervisor, working within those groups were recruited for participation. In addition to Hackman's (1990, pp. 3-5) defining properties of work groups, other constraints were necessary to fit within the theoretical framework of this study: 1) work groups members had to be colocated within the same building; 2) groups had to be internally-oriented/non-customer service; 3) groups had to have one supervisor per group; 4) groups had to range in size from three to 15 members; and 5) groups had to be intact for at least six months. Nineteen groups were sampled, ranging in size from three to fourteen employees. A total of 138 employees were recruited for participation. Of these, 105 participated, culminating in an overall response rate of 76%. In terms of response rate by group, two groups had less than half of the members responding, with rates of 38% and 40%. The other seventeen groups aU had a majority of members responding, with rates ranging from 60% to 100%. Figure 1 Conceptual Model with Hypothesized Relationships of Communication Properties Mediating Physical and Social Factors' Impact on Feelings of Belonging Physical Environment Factors Density Openness Social Climate Factors Group Cohesion D c:::::> c::> Hypothesis Number Group-Ievel Individual-Ievel ... Mediating Relationships - -.... Direct Relationships refereed papers . 73 The nineteen groups were selected from four organizations. Eight were laboratory groups at a large Midwestern university; six were bio-medical groups; and two were engineering groups. Seven were administrative and operations groups in the administrative headquarters of a small Midwestern bank. Two were operations and consulting groups in a Midwestern health care information company. Two were technology groups in a Midwestern information technology consulting company. Design This study used field research methodology at multiple research sites. The sample of nineteen groups was a convenience sample with a minimum of two groups from every participating organization. Sorne groups were recruited by the snowball sampling method (Henry, 1990, p. 21), such that recruited participants recommended other eligible participants for inclusion. Physical measurements and survey (self-administered questionnaire) methods were used to colle ct the data. Measures Physical environment factors were calculated based on floorplan measurements. Group work space density was calculated by dividing the square footage for the group's space by the number of employees in the work group. Openness was calculated by dividing the total square footage of the group work space by the totallength of its interior waUs. For the interpersonal communication constructs, respondents were asked questions for each member of their work group. Dyad-Ievel values were computed, and then the total responses of one individual about aIl ofhis/her co-workers were averaged together. To assess intensity of ties, respondents were asked to "estimate the total time (in hours) spent in face-to-face communication for a typical work week:' To assess enjoyment of face-to-face communication, respondents were asked to "rate the degree to which you enjoy communicating in general with this person:' Other constructs were measured with various questionnaire indices. Group cohesion (a = .89) was measured with an index consisting of 5 Likert -type 74 . edra 29/1998 items from Koys and DeCotiis' Cohesion Dimension of Psychological Climate scale (1991). Supervisor support (a = .90) was measured with an index consisting of 5 Likert-type items from Koys and DeCotiis' Support Dimension of Psychological Climate scale (1991). For the two social climate constructs, after each index was formed, individual responses were aggregated within each group to form group-Ievel measures. To assess feelings ofbelonging, a subscale (slightly modified) from the de JongGierveld Rasch-Type Scale (de Jong-Gierveld & Kamphuis, 1985) was used (a = .91). Two control variables were also measured with indices: introversion/extroversion (from the Keirsey Temperament Sorter of the Myers-Briggs Type Indicator) and communication apprehension (from McCroskey's (1977) Personal Report of Communication Apprehension - 24 index). Results Preliminary statistics were run including descriptive univariates and bivariate correlations. lndividual hypotheses were tested with the hierarchical linear modeling (HLM) technique (Bryk & Raudenbush, 1992), and the overaU conceptual model was tested with path analysis Ce.g., Blalock, 1971). lndividual Hypotheses To test hypotheses 1 through 4 (direct effects), a cross-Ievel regression was conducted with aU four group-Ievel exogenous variables. To produce the results, an HLM means-as-outcomes regression model was run. To test hypotheses 5 through 14 (mediating relationships), cross-Ievel regressions were run for the three endogenous variables. To regress intensity of ties on the four group-Ievel predictors and enjoyment of communication on the four group-Ievel predictors, the HLM means-asoutcomes regression models were used as described above. To regress feelings of belonging on the two mediating individual-Ievel variables, the square mot of intensity of ties and enjoyment of communication, and the two control variables, introversion/ extroversion and communication apprehension, the HLM random-coefficients regression model was used. Details of the HLM technique are described in Bryk and Raudenbush (1992). Flgure 2 . ., !i~ediational Model of Communication Properties Mediating Physical and Social Factors' Impact on Feelings of Belongmg wlth Results Density Openness .66 Intensity ofTies Enjoyment of Communication Group Cohesion Introversion/ Extroversion <:::) Group-Ievel c:> Individual-Ievel Due to the small sample of groups (N = 19), the statistical tests had low power, and many relationships were not found to be statistically significant, while there may have in fact been true relationships. Many of the results were in the expected direction of their respective hypotheses, but they were not statistically significant at the standard test level of p < .05. The following hypotheses were those from the set of fourteen that were statistically supported. • Hypothesis 3: Employees in groups with greater group cohesion have stronger feelings ofbelonging than those in groups with less group cohesion (g = .39, t = 2.48, P < .05). • Hypothesis 11: Employees in groups with greater group cohesion enjoy communicating with their coworkers more th an those in groups with less group cohesion (g = .43, t = 2.89, P < .05). • Hypothesis 14: Employees who enjoy communicating with their co-workers have stronger feelings ofbelonging than those who do not enjoy communicating with their co-workers (g = .44, t = 4.95, P < .001). * p<.10 ** p < .05 Overall Model From the results of the cross-Ievel regressions, a causal path model could be formed with four exogenous group-Ievel variables and three endogenous individual-Ievel variables. The model with results is provided in Figure 2. The predictors vary to the degree they explain variance in the endogenous variables: the four group-level predictors explain 34% of the between-group variance in the square root of intensity of ties; the four group-Ievel predictors explain 73% of the between-group variance in enjoyment of communication; and the four individual-level predictors explain 18% of the withingroup variance in feelings of belonging. To test the overall fit of the model, Q and W values can be computed and tested with the chi-square statistic as described in Pedhazur (1982, pp. 617-620). The current data led to the results that Q =.713 and W ( 4) = 34.216, based on a sample size of 105. The critical value of c2 for df = 4 is 9.488, so W is significant and the model is rejected based on the c2 test. Comparing the value of Q to 1.0, the model is not a poor fit of the data, but is not an excellent fit either. refereed papers . 75 To determine if the relationships investigated are truly mediational, it is important to calculate not only the indirect effects of the four group-Ievel variables on feelings of belonging as was done above, but to also test if the direct effects are eliminated when the two mediators are included in the regression (Baron & Kenny, 1986). If intensity of ties and enjoyment of communication are full mediators, then the four direct relationships should disappear when these two communication variables are entered into the regression (Baron & Kenny, 1986). Although not entirely eliminated, three of the direct relationships were reduced as follows: group cohesion's effect decreased from g = 0.390 to g = 0.154; supervisor support's effect decreased from g =0.180 to g =0.118; and openness decreased from g = 0.021 to g = 0.005. The only direct relationship which was not reduced by the addition of the two communication variables was density, which increased, indicating a suppression effect (Pedhazur, 1982,pp.104-105). Thus, the relationships between group cohesion, supervisor support, and openness with feelings of belonging were partially mediated by intensity of ties and enjoyment of communication. Discussion The first purpose of this study was to address the research question: How do group social climate factors in conjunction with group physical environment properties predict individual feelings of belonging? When examined concurrently, the social climate factors of group cohesion and supervisor support were more strongly related to individual feelings of belonging than the physical properties of density and openness. Among the four grouplevel antecedents, group cohesion had the strongest influence and density had the weakest. Of the four, only group cohesion had a statistically significant relationship with feelings ofbelonging. Together, the four antecedents explained about 66% of the variance between groups in employees' feelings of belonging. These fmdings indicate that when the social climate and the physical environment of group work settings are considered simultaneously, the social cli- 76 . edra 29/1998 mate has more of an impact on employees' feelings of belonging. In the past, there has been little rese arch on how physical and social environments together affect interpersonal interactions and individual outcomes in work settings. When examined concurrently, group cohesion and supervisor support have more of an impact on individuals' feelings of belonging than the density or openness of the group's work space. So although properties of the physical environment may be important in determining certain interpersonal and individual outcornes in the work setting, when predicting whether group members will feel a sense of belonging towards their group, the physical features of their work settings are not as important as the social climate. In particular, it is the cohesiveness of the group that best predicts if employees will feel a sense ofbelonging. The second component of the research was: Does interpersonal communication mediate the relationships between these group properties and individual feelings ofbelonging? For the most part, the answer is no. The largest indirect effect, with intensity of ties and enjoyment of communication serving as mediators, was the indirect relationship between group cohesion and feelings ofbelonging, and it was fairly small. The next largest indirect relationship was between density and feelings of belonging, though it is most likely a result of density acting as a suppressor variable since the strength of the direct relationship between the two was much smaller, so no substantive conclusions should be drawn. For both openness and supervisor support, their relationships with feelings of belonging were not mediated by the two communication properties. The overall conclusions drawn from these findings are that (1) interpersonal communication properties have no bearing on relationships between density, openness, and supervisor support on the one hand, and feelings of belonging on the other; and (2) interpersonal communication partially mediates the relationship between group cohesion and feelings ofbelonging. In particular, it is employees' enjoyment of communication with co-workers that affects this relationship. So group cohesion has both a direct effect on individuals' feelings of belonging and also an indirect effect through group members' enjoyment of communication with their co-workers. Limitations of this study include the following. Since the interpersonal communication constructs were measured with self-report data, they may have been biased, since participants may under - or over-report true amounts due to poor recall or estimation. Since nonprobability sampling procedures were used, the findings may not be valid due to selection bias, and caution should be used before generalizing. Due to the small sample sizes, it is unclear if there were truly no relationships in sorne cases, or just too little statistical power to detect them. Though the conceptuaI model is recursive, in reality, for many of the relationships, the causal flow probably goes in both directions. Finally, since the causal links asserted in the conceptual model could not be tested (since this study used a cross-sectional field setting design), only correlation al conclusions can actually be drawn. Before pursuing future theoretical development, the initial theories that generated the current study must be reexamined. Turning to the theoretical frameworks (Becker, 1981; Moos, 1976) that guided the conceptual mode!, an important question is, do the current results support their models? Becker (1981) proposed that physical features such as density and arrangement of work settings indirectly affect individual outcomes as mediated by both the qualityand level of interpersonal communication. Based on this study's findings, Becker's assertions were not supported. Becker may have not implicitly included feelings of belonging in his targeted group of individual outcomes. The three outcomes listed in his conceptual model were turnover, personnel, and absenteeism. So, although it is not clear whether the currently findings are directly applicable to the specifics of his model or not, at least in spirit, they do not offer support for it. sor) support were examined. As with Becker's model, it is not clear if Moos implicitly included feelings ofbelonging as a potentially affected outcome, though it is feasible that his notion of morale might be conceptually similar. This study's findings did offer support for the aspects of Moos' model that were investigated. Group cohesion, and to a lesser degree, supervisor support did directly impact individuals' feelings of belonging. Turning to future the ory development, researchers such as Moos (1976) , Walsh (1987), and Stokols (1990) have called for more research that combines the study of physical as weIl as social contexts as they impact interpersonal pro cesses and individual behaviors and attitudes. Further theoretical work is needed to understand why the physical environment affects intensity of ties but not enjoyment of communication and feelings ofbelonging, and why the social climate affects enjoyment of communication and feelings of belonging but not intensity of ties. AIso, theory is needed not only to guide research on cross-Ievel contextual effects, but also regarding how group-Ievel social and physical properties affect each other. As demonstrated in this study, physical and social properties of work groups were considerably related, but the reasons why remain unclear. Clearly, further theoretical development is needed to guide future research in this area of study. References Alderfer, c.P. (1972) Existence, Relatedness, and Growth: Human Needs in OrganizationalSettings. New York: Pree Press. Altman, 1. (1975) The Environment and Social Behavior. Monterey, CA: Brooks/Cole. Anant, S.S. (1970) Belongingness and Socialization: A Developmental Theory of Belongingness. Psychologia: An International Journal of Psychology in the Orient, 13(2-3),102-112. Moos (1976) offered a model in which the social relationship dimensions of involvement, peer cohesion, and staff support in the workplace are supposed to impact individual attitudes and behaviors such as satisfaction, morale, absenteeism, and turn0ver (see Figure 2). In the current study, the dimensions of peer (group) cohesion and staff (supervi- Barker, R.G. (1968) Ec%gical Psych%gy: Concepts and Meth- ods for Studying the Environment of Human Behavior. Stanford, CA: Stanford University Press. Baron, R.M., and Kenny, D.A. (1986) The Moderator-mediator Variable Distinction in Social Psychology Research: Conceptual, Strategie, and Statistical Considerations. Journal of Personality and Social Psych%gy, 51,1173-1182. refereed papers . 77 Baumeister, R.F., and Tice, D.M. (1990) Anxiety and Social Exclusion. Journal of Social and Clinical Psychology, 9(2),165-195. Stokols,D. (1990) Instrumental and Spiritual Views of Peopleenvironment Relations. American Psychologist, 45(5),641-646. Becker, F.D. (1981) Workspace: Creating Environment in Organizations. New York: Praeger. Walsh, W.B. (1987) Person-environment Congruence: A Response to the Moos Perspective.fournal ofVocational Behavior, 31(3),347-352. Blalock, H.M., Jr. (1971) Causal Models in the Social Sciences. Chicago: Aldine-Atherton. Bryk, A.S., and Raudenbush, S.w. (1992) Hierarchical Linear Models: Applications and Data Analysis Methods. Newbury Park, CA: Sage. de Jong-Gierve!d, J., and Kamphuis, F. (1985) the Deve!opment of a Rasch-type Lone!iness Scale.Applied Psychological Measurement, 9(3),289-299. Fromm, E. (1968) Revolution of Hope. New York: Harper and Row. Hackman, J.R. (1990) Groups that Work (and those that don't). San Francisco: Jossey-Bass. Henry, G.T. (1990) Practical Sampling. Newbury Park, CA: Sage. Herzberg, F., Mausner, B., and Synderman, B.1959. The Motivation to Work. New York: John Wiley and Sons. Klein, K.J., Dansereau, F., and Hall, R.J. (1994) Leve!s Issues in Theory Development, Data Collection, and Analysis. Academy ofManagement Review, 19(2),195-229. Koys, D.J., and DeCotiis, T.A. (1991) Inductive Measures of Psychological Climate. Human Relations, 44(3),265285. Maslow, A.H. (1954) Motivation and Personality. New York: Harper and Row. McClellend, D.C. (1971) Motivational Trends in Society. Morris-town, NJ: General Learning Press. McCroskey, J.e. (1977) Oral Communication Apprehension: A Summary of Recent Theory and Research. Human Communication Research, 4(1),78-96. Monge, P.R., Edwards, J.A., and Kirste, K.K. (1983) Determinants of Communication Network Involvement: Connectedness and Integration. Group and Organization Studies, 8(1), 83-11l. Moos, R.H. (1976) The Human Context: Environmental Determinants ofBehavior. New York: John Wiley and Sons. Oldham, G.R., and Rotchford, N.L. (1983) Re!ationships between Office Characteristics and Employee Reactions: A Study of the Physical Environment. Administrative Science Quarterly, 28,542-556. Pedhazur, E.J. (1982) Multiple Regression in Behavioral Research: Explanation and Prediction (2nd ed.). Fort Worth, TX: Harcourt Brace. Rousseau, D.M. (1985) Issues of Leve! in Organizational Research: Multi-leve! and Cross-leve! Perspectives. In L. Cum-mings and B. Staw (Eds.),Research in Organizational Behavior. Greenwich, CT: JAl Press, 1-37. 78 . edra 29/1998 Wicker,A.W. (1992) Making Sense of Environments. In W.B. Walsh, K.J. Craik, and R.H. Price (Eds.), Person-en- vironment Psychology: Models and Perspectives. Hillsdale, NJ: Lawrence-Erlbaum Associates, 157192. Public Policy and Mass Housing in Brazil Antônio 1. Reis and Maria C. Lay University of Rio Grande do Sul, School of Architecture This paper looks at mass housing stock in Brazil after 1964, when a national housing policy was adopted and the National Housing Bank (BNH) was created. Therefore, the intersection ofenvironment and behavior and public policy is examined with respect to mass housing in Brazil within that context. The paper examines the main socioeconomic and physical characteristics of mass housing produced by the BNH; it further explores the relationship between public policy and the mass housing produced and identifies the main factors affecting user satisfaction and behavior. Results from a research carried out in three housing estates comprisingfour different dwelling types is presented together with results from other studies. Data was collected by means of questionnaires, interviews and observations. It was revealed that there are several negative aspects associated to quantity and quality in the housing provision yet to overcome. Just as political action must be taken in order to increase investments in housing provision and to improve the socio-economic situation of the large lower-income groups of the Brazilian population, also the shape and spatial structure of the subsidized housing provided in Brazil need to be changed according to the design factors so far identified as affecting performance, sa that significant improvement can be achieved in mass housing design. Introduction This paper looks at mass housing in Brazil after 1964, when a national housing policy was adopted and the National Housing Bank (BNH) was created. Therefore, the intersection of environment and behavior and public poliey is examined with respect to mass housing in Brazil within that context. A brief history of two main related aspects of 1964 mass housing policy is presented. The paper examines the main socioeconomie and physical characteristics of mass housing produced by the BNH; it further ex- plores the relationship between publie policy and the mass housing produced and identifies the main factors affecting user satisfaction and behavior. Two Main Related Aspects of 1964 Mass Housing Policy: Migration and Population Growth The urban population growth has been the main cause of a huge increase in the number of housing units needed in the urban are as, since the last four decades, caused mainly by two factors: migration from rural areas and the significant increase in the total Brazilian population. After 1950,Brazil went through a chaotic urbanization, due to an uncontrolled migration from the rural areas toward the urban are as, mainly caused bya quick industrialization (Pedreira, 1985).As industrialization did not follow any integrated development poliey, profound imbalances were generated, amplifying the regional differences and promoting desertion of the countryside. As a result of the intense migration movement, the urban centers had a phenomenal expansion, creating physical and social effects (Guglielmi, 1985). By the end of the 1950s, the urban population was 70% larger th an it was at the beginning of that decade. As this increase was maintained during the 60s, the growth pro cess produced a severe shortage of urban amenities (Serra, 1989).Amongthemanyshortages (such as infrastructure, community facilities, and publie serviees), the housing one was highlighted, with the proliferation of'corti,cos' (sIum tenements), squatter settlements and precarious self-built houses on the urban periphery (Guglielmi, 1985). Also contributing to the urban explosion was the explosive rate of population growth . From 1950 to 1980 the total Brazilian population increased from 52 to 119 million, an increase of 128% in 30 years. During the same period the urban population increased from less than 19 million to more than 80 million, growing by 321 %. The urban population refereed pa pers . 79 was 45% of the total in 1960 and 67% of the total by 1980. Today, with an estimated population of 150 million people, around 112 million are calculated to be living in urban are as and about 38 million in rural areas. This trend indicated that 75% of the Brazilian population will live in urban areas by the end of this century, at which time the total population would be more than 165 million. However, the number of inhabitants in rural areas at the end of this century would be the same Brazil had in 1950 (Serra, 1989). Despite the comments on that urbanization cufve of Brazilian population was beginning to descend, so that the impact of migration fluxes on urban life ought to be smaller each time, as well as the demographiegrowth was also diminishing (Singer, 1985), it, however, does not alleviate the seriousness of the housing number and quality-related problems, as shown in the next sections. Mass Housing after 1964: Public Policy and Socioeconomic Related Aspects Only in August 1964, after the military coup, was the first national housing policy elaborated, with the creation of the BNH (National Housing Bank), the CNDU (National Commission for Urban Development) and the SFH (Housing Finance System) (Maffei, 1974; Batley, 1983). The BNHwasto'finance publie housing and urban infrastructure: according to SFH finance poliey, using money from compulsory savings-the FGTS (Guarantee Fund for Employment Time), whieh was a kind of unemployment insurance-and from voluntary savings in banks or finance institutions. The CNDU had the task of'preparing plans and proposing laws dealing with the explosive urban growth' (Serra, 1989, p.309). Moreover, it was remarked that the military coup of March 1964 needed, in order to sustain its government, a poliey of popular involvement and economie development that generated incomes and jobs, and at the same time showed the sensibility of the new government to housing problems. This justified the creation of BNH and the setting up of a saving policy enabling the creation of a housing system that would give support to other gov- 80 . edra 29/1998 ernment initiatives of the 'revolution'. As a consequence, the number of units financed increased from 29 thousand in 1966 to 148 thousand in 1968. This demonstrated the relationship between housing policy and the political and economic necessities of the government (Guglielmi, 1985). These points are referred to by Darke and Darke when discussing the reasons for state intervention in housing (Darke and Darke, 1988, pp.4445). The BNH-financed mass housing and infrastructures was promoted byCOHABs (municipal or state housing company), among others, and generally produced housing for people with incomes varying between three and five minimum salary. This was in practice the lowest income band acceptable for those applying for a dwelling in one of the housing estates financed by the BNH; this was called "low income;' "popular;' or "economic" housing (Batley, 1983, p.75). This does not mean, however, that low-income people were the ones to benefit most from BNH housing policy. Although the main purpose of the BNH was said to provide housing for low-income people, most of the people housed were low-middle income people, because these were the people who could afford to pay for these dwellings (Projeto 1987, Batley, 1983; Marieato, 1987). Singer pointed out that the creation ofBNH was a considerable move in housing poliey. He argued that its inspiration was possibly the liberal North -American model of savings and loans associations that avoided the direct intervention of public funds. However, its application in Brazil was considerably distorted since aIl the FGTS money was applied to schemes creating mass housing but actually not occupied by lower incomes, yet financed by the workers' own salaries. The FGTS was in fact used to feed the "boom" of the building industry that was orientated to the middle classes, whose real income the "economic miracle" (significant economic growth between 1968 and 1973) expanded in a notable way (Singer, 1985). After the Brazilian "economic miracle" it was riecessary to recover the prestige lost with the working class. During the succeeding government, the salary squeeze was alleviated and a bigger portion of BNH funds were applied to housing that could be considered "popular" or "social;' though it rarely reached the poorest (Singer, 1985). Whereas during the so called "economic miracle" an ever-Iow number ofhousing units, 54 thousand units in 1972, were financed by the BNH, after the "economic miracle" 249 thousand units in 1977 and 339 in 1979 were fmanced by the BNH. This was an apparent demonstration of change of governmental position after the defeat in the regional elections of 1974 (Guglielmi, 1985). During the "political openness" of successive military governments, an audacious massive housing production policy was established, with 626 thousand units financed in 1980 and 560 thousand units in 1982, with the objective of getting approval from the working class sectors. Though the housing production from 1980 allowed delay of an economic crisis for more three years, the recession and market retraction felt in 1983, with the arrivaI of the IMF, generated a fmancial collapse that was reflected not only in the building industry but in the national economic performance as a whole. This situation was not altered in the government of'Nova Republicà (New Republic - from 1985 to 1990), the first civil government after 21 years of military dictatorship. The level of unemployment continued to be high, incomes were kept low, and the discrepancy between mortgages and readjusted salaries did not stop. As a consequence four and a half million borrowers were declared defaulters (Guglielmi,1985). The BNH and SFH were extinguished in 1986 and resources and projects transferred to the CEF (Caixa Economica Federal- Federal Saving Bank). During its 23 years of existence, the BNH did not produce more than four and a half million housing units. of which one million and six hundred thousands units were 'popular' or 'economic' dwelling units promoted by COHABs income band. This left an estimated deficit of 10 million housing units, in addition to the growth of squatter settlements and sIums (Pedreira, 1985;Veja, 1991). cal characteristics of the housing estates that were built. Generally these mass housing schemes reached 500 units, housing about 2000 to 2500 people, though they could reach 3000 or 5000 units or even 20000 units, with about 100,000 residents or even 200,000, as in Cajazeira housing estate on the periphery of Salvador City in northeast Brazil (Pedreira, 1985; Veja, 1991). Housing estates in Sao Paulo tended to be the largest ones, and might reach the dimensions of a medium sized city of 150,000 inhabitants, as in the case of Itaquera 1,11, and III (Maricato, 1987). Figure 1 Figure 2 Four Storey Blocks of Flats Oetached Houses in Small Plots Figure 3 Figure 4 Two Storey Blocks of Flats Row Houses Mass Housing after 1964: Physical Characteristics, Dwelling TYpes and Layouts The housing estates have been mainly characterized by two dwelling types: the repetitive block of flats, usually without lifts and four storeys high (Figure 1); and the detached houses on small individual plots (Figure 2). While the block of flats followed the modernist approach with blocks placed in continuous open space without differentiation and not related to the street, the detached houses on small plots followed the conventional division of plots with "a narrow and long block arrangement. Two storey blocks of flats (Figure 3) and row houses (Figure 4) were also built. The problems of mass housing in Brazil were related not only to the number of housing units produced and the fact that the poorest were excluded from the pro cess ofhousing allocation, but also to the physi- Generally, the housing estates were located on the urban periphery, in old urban empty plots with large dimensions, or built many kilometers from the city center, expanding the borders of the city (Herrmann, refereed pa pers . 81 1985; Pedreira, 1985). The land in-between was transformed from rural to urban plots, the object of severe speculation (Herrmann, 1985). Sorne housing estates had a verypoor building quality. Without any environmental comfort, these estates were created only as a dormitory, practically without leisure, health, and educational facilities, telephone services or public transport (see Pedreira, 1985). AIso, there was a lack of adequacy of building materials to climatic conditions in many cas~s, and it was uncommon to use regional technologIes (Souza in Almeida, 1985). Brazil did the contrary of Colombia, where a variety of proposaIs involving the private sector and popular movements in different combinations were experimented with. The Brazilian government adopted, during the same time, indistinguishable official projects that covered the peripheries of medium and big cities with bad examples of urban design and architecture for more th an 20 years (Santos, 1988a). Mass Housing after 1964: Pro cess of Production and Allocation of Housing Units Norrnally, the buyers of the housing units built by the COHABs did not even know where they were going to live, even less if their housing units would be houses or flats, or if built with blocks, concrete walls or any other building material. The general decisions ofwhy,how, where and when the housing estate should be built were determined by the institut ions involved with its promotion and finance, which had with political and economic ambitions different from those that would be defended by the future users. Moreover, the BNH did not allow for a housing policy that permitted quality control of the aIl processes of production and for sorne sort of user participation as weIl since aIl the projects followed a nationwide orientation dictated by the BNH[, which controIled the resources and the release of funds (Souza in Almeida, 1985). Hence, the total dominance of the producer viewpoint over the user viewpoint during aIl the production pro cesses including the design, in housing provision in Brazil was clear and irrefutable. 82 . edra 29/1998 The Problem of User Dissatisfaction with Mass Housing in Brazil Although, when the housing deficit is very significant, it is easy to emphasize quantitative aspects rather than qualitative ones, these do not lose importance. It seems weIl established that the importance of thc qualitative aspects concerning mass housing in Brazil has been neglected. Those housing estates and dwelling units that have been produced have not satisfied their users in many respects. It was properly reported that many Brazilian housing estates are bad, not necessarily because they are cheap, but because they are badly planned, badly provided with services and facilities, and built with the classic lack of governmental care when dealing with low in come people (Veja, 1991). It is shown in the next sections various aspects of the housing environment that have not satisfied its users. These aspects are, mainly, the result of the research carried out in a number ofhousing estates in southern Brazil, comprising four different dwelling types, namely, detached and row houses, two and four storey blocks of flats (figures 14). Questionnaires, structured interviews, observations ofbehavior and physical traces were used as data gathering means (Reis, 1992, Lay, 1992). Dwelling Related Factors Affecting User Satisfaction There were a number of common causes for dissatisfaction to all dwelling types related to rooms, either size, layout or number. The poor quality of materials and workmanship were also a problem for most except for those living in row houses. Lack of habitability of the housing units, deterioration of the materials, and plumbing not working properlywere also problems found in the dweIling units (Souza, 1985). Moreover, many technical problems originated in the inadequacy of projects and materials used in certain regions and sites. Similar projects were used in regions ofBrazil with different climatic conditions and in sites with different topography. As an example, it can be mentioned the use of concrete blocks without external pl aster in the Rubem Berta housing estate in Porto Alegre, which is not proper for its humid climate; rain water has infiltrated (Reis, 1992). The case of the Santa Etelvina Table 1 Main Factors Affecting Satisfaction with the Dwelling DwellingType % Main Factors Provo king Dissatisfaction % Main Factors Inducing Satisfaction % Main Factors Users would Like to Improve 38.4 21.4 11.9 11.9 11.9 11.9 11.9 increase dwelling size change the size of rooms fence around patio change room use/fune. change pos. of door/window change/imp. materials/work. security from burglar Detached Houses Tancredo 35.7 rooms arrangement 26.2 quality of materials/work. 16.7 number of rooms Detached Houses Costa 60.0 30.0 20.0 20.0 20.0 20.0 number of rooms size of rooms quality of materials/work. position of doors/windows lack privacy in the dwelling size of patio 20.0 location of house in esta te 70.0 change the size of rooms 30.0 change room use/function 30.0 change/imp. materials/work. Row Houses Costa 59.0 50.0 45.4 36.3 31.8 row houses themselves number of rooms type of access in backyard size of backyard size of rooms 27.2 location of house in estate 22.7 qua lit yof materials/work. 63.6 40.9 27.2 27.2 27.2 change the size of rooms inerease dwelling size change room use/function type of access backyard change dwelling type itself Two Storey Blocks of Flats Taneredo 65.9 size of rooms 26.8 quality of materials/work. 14.6 nothing they don't like nothing they like location of fiat in estate quality of materials/work. security from burglar 58.5 22.0 22.0 17.1 12.2 change the size of rooms change/imp. materials/work. have a patio or balcony increase dwelling size fence around the block Four Storey Blocks of Flats Rubem Berta 65.0 40.0 30.0 27.5 20.0 17.5 12.5 55.0 37.5 32.5 27.5 20.0 20.0 15.0 12.5 change/imp. materials/work. external block appearance add/remove partition wall internai privacy change the size of rooms room appearance change room use/function increase dwellin 9 size quality of materials/work. rooms arrangement lack privacy in the dwelling humidity wind infiltration water/sew. elecJtelephone number of rooms nothing they don't like 23.8 14.3 11.9 11.9 24.4 14.6 12.2 12.2 almost everything qua lit yof materials/work. location of house in estate everything 45.0 nothing they like observation:frequencies were obtained from open questions,are not mutually exclusive,and are ranked by descending order. refereed pa pers ·83 housing estate in Sao Paulo is also worth mentioning, where the many repairs mainly due to topographic problems put up the original cost of PUS 6000 dollars a flat to about PUS 18000 dollars, equivalent to the cost of accommodation with the same dimensions in a better neighborhood (Veja, 1991). Other reasons given for dissatisfaction were related to sorne aspects of design within th~ house such as lack of privacy (detached Costa, four storey Rubem) and other particular faults (see Table 1). Lack of visual and auraI privacy were problems also established in the work done by IPT. Moreover, household size (density) was very important in affecting resident satisfaction, principally in relation to privacy and cariying out activities (Motta and Del Carlo, 1975). None of these problems, it is argued, are related to a fundamental characteristic of a dwelling type-they are design faults that could be found in any type. However, two of the reasons for dissatisfaction given by those living in row houses were a function of their type-a criticism of the type itself characterized by the common side walls, and the lack of direct external access to the backyard only found in row houses. Size of dwelling and rooms was the factor that residents most wished to change in all types of dwelling, though to a lesser extent in the four storey flats. In these there were many other concerns that predominated related to quality of the flats, their design for the point of view of privacy and also overall appearance. In detail, table l indicates different priorities and differences in the wish to change aspects of their dwellings. Those factors causing dissatisfaction were, in general, all mentioned as needing to be improved. In research carried out by COHAB of Sao Paulo, in Presidente Castelo Branco housing estate, it was found that though sorne 65% were said to be satisfied with the size of their dwellings, 78% said the y aspired to make changes in the future. These changes were not, however, related to major needs of user expression or personalization, but mainly related to increasing the size of rooms (52%), addingone more room to the house (29%), and to general improvements such as roof, ceiling, floor, plas- 84 . edra 29/1998 ter the walls, change of doors and windows, and paint in general (21 %), and separation of living room and kitchen (9%). Nonetheless, financial difficulties of the residents in spending money on the maintenance of the dwellings was noted. Only 22% were said to be satisfied with the dwelling, followed by Il % who were said to be satisfied with the living room (used by many as living room during the day and as bedroom during the night), and Il % who were said to be satisfied with the location of the dwelling in the estate and the location of the estate itself. The factors most provoking dissatisfaction were noise, mainly from children playing in corridors and staircases in blocks of flats due to lack of open spaces and play are as, noise from pets in houses or flats and the consequent effect on user interaction (15%), poor quality of materials and workmanship (10%), and lack of window shutters (9%) (COHAB-SP, 1978). It is worth noting that though the majority were said to be satisfied with the size of the dwelling, only a minority mentioned the dwelling as a whole as promoting satisfaction. It might be that users were not actually satisfied with the dwelling size, as the des ire to increase room size indicated, but that the majority answered positively, perhaps because this question was put right at the beginning of the questionnaire, and the questionnaire was administered by the CO HAB itself. This might have had sorne inhibiting effect on user answers, and might also be related to the rather conspicuously low percentage of residents mentioning factors provoking dissatisfaction, such as quality of materials and worlunanship. Housing Es ta te ReZated Factors Affecting User Satisfaction It is shown in Table 2, that play spaces for children is a problem in aIl housing estates, especiaIly in Costa, and were clearly an important aspect to be improved in all estates. Moreover, a confrontation between the estate layout and user interests has been detected through the under-utilization or abandonment of collective areas and community facilities provided (Lay, 1992; Reis, 1992; Carvalho, 1985). Those in four storey blocks in Rubem Berta were clearly the least satisfied with a high proportion Table 2 Main Housing Estate Factors Affecting Satisfaction by Dwelling Type DwellingType % Main Housing Estate Factors Provo king Dissatisfaction % Main Housing Estate Factors Promoting Satisfaction Main Housing Estate Factors Users would Like to Improve % Detached Houses Ta ncredo 35.7 21.4 11.9 11.9 11.9 nothing they don't like paved streets/footpaths play space for children health centerfacility shops/supermarket 45.2 38.1 16.7 11.9 11.9 shops public transport health center facility school drugstore 26.2 23.8 19.0 19.0 19.0 11.9 11.9 paved streets/footpaths play spaces for children health center fa cil ity shops/supermarket nothing to improve schools security from thieves Detached Houses Costa 30.0 30.0 20.0 20.0 20.0 20.0 play spa ces for children security from vandalism estate layout esta te maintenance estate c1eanliness nothing they don't like 30.0 20.0 20.0 20.0 public transport paved streets/footpaths street lighting school 50.0 play spaces for children 20.0 security from thieves 20.0 nothing to improve Row Houses Costa 31.8 31.8 27.2 22.7 play spaces for children nothing they don't like security from vandalism estate maintenance 22.7 public transport TwoStorey Blocks of Flats Tancredo 29.3 19.5 19.5 17.1 14.6 12.2 12.2 paved streets/footpaths play space for children electric/water sewage system nothing they don't like estate c1eanliness public telephone exterior estate appearance 22.0 22.0 17.1 17.1 shops/supermarket nothing they like public transport sense of community 26.8 19.5 17.0 14.6 12.2 12.2 12.2 fence around the block paved streets/footpaths play space for children trees/greenery change/imp. materials/work. esta te col ors exterior estate appearance Four Storey Blocks of Flats Rubem Berta 45.0 45.0 42.5 17.5 15.0 15.0 15.0 rubbish collection estate c1eanliness paved streets/footpaths how open spa ces are used play spa ces for children organization users in a block security from thieves 37.5 20.0 20.0 15.0 12.5 nothing they like public transport. location of estate sense of community shops/supermarket 35.0 27.5 15.0 12.5 12.5 12.5 12.5 12.5 paved streets/footpaths change/imp. materials/work. child's play spaces use of open spaces estate c1eanliness organization users in a block security from thieves securi ty for child's pla yg round 31.8 security from thieves 22.7 play space for children observation:frequencies were obtained fram open questions,are not mutually exclusive,and are ranked by descending order. refereed papers . 85 users mentioning poor rubbish collection and lack ofhousing estate cleanliness among many other factors causing dissatisfaction. In addition, the improvement of pavement of streets and of footpaths was a major concern of those in detached and two storey blocks in Tancredo, and in four storey blocks in Rubem Berta estate. A lack of conservation of collective areas, of lateral walls of buildings, and of services and pavements, has also been found through observations (Reis, 1992; L<\y, 1992; Carvalho, 1985). It seems worth noting that between the two factors ranked first among those provo king dissatisfaction in the four storey blocks in Rubem Berta, namely rubbish collection or refuse disposaI facilities, and estate/neighborhood cleanliness, the former was not even directly mentioned and the later came only as the fourth factor to be improved together with other four. This could be due to respondents option to expend money improving pavements, streets and footpaths, while living rubbish collection and estate/neighborhood deanliness to be improved bypromoting organization and education among the residents. In general, however, it could be said that a relationship existed between what the users indicated as causing dis satisfaction and what they would like to improve in their housing estates (Table 2). Shops in the Tancredo estate was liked by the residents, particularly by residents of the detached houses (Table 2) because shops were more integrated in the houses area or/and because those in fIats were generally less enthusiastic about the Tancredo estate. Nonetheless, a tendency to isolate housing from the other functions generaIly present in spontaneous settlements was noted. Planned in a hurry,most ofhousing estates promoted by COHABs do not have a shopping area in the original project. Shops appear later, not always in the most appropriate places. Garages were used, and shops built in the common open spaces (Veja, 1991). This was also the case in Rubem Berta housing estate (Reis, 1992). Security from thieves in the streets not mentioned only by those in two storey blocks in Tancredo, was a main factor to be improved according to those in row houses in Costa. In two storey blocks of fIats in Tancredo, having a dear marking (fence) between the semi -pnvate space around the block of fIats and the public space, was the most mentioned factor as 86 . edra 29/1998 needing to be improved, though its absence was not mentioned as a cause of dis satisfaction (Table 2). It has been mentioned that poor housing estate location, and layout not suitable to the physical environment were among the estate-related problems (i.e. Souza, 1985). Maricato mentioned that the estates were built far away from urbanized areas, mainly because of the low co st of land (Maricato, 1987). Moreover, Santos commented that the housing estates, distant and isolated, were revealed as equally or more problematic than the settlements theywere supposed to replace (Santos, 1988). It can be seen in Table 2 and in COHAB's findings shown below, however, that these daims are not entirely supported. It might be the case that the inconvenience of transport and distance to workplace could be compensated bya rather better physical environment, in terms of green space than that found in more urbanized areas in the city. For example, in the research carried out by COHAB in the Presidente Castelo Branco housing estate the factors most promoting satisfaction were green areas in the estate (25%), and the location of the estate, mainly because of its provision of pure air, panoramic views, and tranquillity (24%) (COHAB-SP, 1978). Moreover, sin ce the provision of public transport was an important factor promoting satisfaction in aIl estates, especially valued by those living in detached houses (Table 2), it might be argued that with an adequate system of transport, location by itself it is not an indicator of adequacy in terms of distance to central areas. On the other hand, when asked about satisfaction with transport, it was found that 94% of residents in the Presidente Castelo Branco housing estate were dissatisfied (43% because of the reduced number of buses and trains, 26% because of the price of the ticket, and 25% because of the bus itinerary). Nonetheless, the factors mentioned as those most provoking dissatisfaction with the housing scheme were constant voltage faIl and lack of electricity (28%), lack of a police office (21 %), lack of pavement on streets (19%), poor transport (12%), lack of a medical center (10%), deficient public lighting (9%), and constant lack of water (9%). Yet, sorne said they were not satisfied with anything on the estate (14%) (COHAB-SP, 1978). Conclusion Hence, the intersection of environment and behavior and public policy examined with respect to mass housing in Brazil after 1964, reveals that the match between users' attitudes and behavior and the housmg environment was not adequately fulfilled by the public housing policy, apart from the quantitative aspects of housing production that have not been accomplished. The problems were due to a lack of concern of future users' needs and values, by the government and by those involved in the process of planning, designing and building the housing estates, whose values, in many cases, were quite different from users' values. Moreover, as argued by Santos (1988), it is necessary to abandon the normative rigidity and the centralization of decisions that had characterized housing production in Brazil. Small projects with control bythe local authorities and with respect for the values and needs of the affected population could be the answer. Therefore, just as political action must be taken in order to increase investments in housing provision and to improve the socioeconomic situation of the large lower-income groups of the Brazilian population, also the shape and spatial structure of the subsidized housing provided in Brazil need to be changed according to the design factors so far identified as affecting performance, so that significant improvement can be achieved in mass housing design. Herrman, J. (1985) A cidade contra 0 homem. Arquitetura e Urbanismo, Novembro, pp. 94-97. Lay, M. C. (I 992) . Responsive Site Design: User Environmental Perception and Behaviour. Oxford Polytechnic, Ph.D. Thesis. Maffei, C. (1974) Towards a Housing System for Sao Paulo. England, Newcastle University, M. Phil Thesis. Maricato, E. (1987) PoUtica Habitacional no Regime Militar: do Milagre Brasileiro à crise Econômica. Petrapolis, Rio de Janeiro, Editora Vozes. Motta, C.F.A. and Del Carlo, U. (1975). Conjuntos Habitacionais: Levantamento das Aspiraçoes e Necessidades do Homen. Sao Paulo, IPT/FAPESP. Pedreira, (1985) Um problema de todo mundo. Arquitetura e Urbanismo, Novembro, pp. 66-67. Projeto, (1987) 0 que 0 Habitee 87 propôs para resolver problema habitacional. Projeto, April. 0 Reis, A. (1992) Mass Housing Design, User Participation and Satisfaction. Oxford Polytechnic, Ph.D. Thesis. Santos, C.N.F. (1988a) 0 futuro esta ai - seremos todos agentes de Nostradamus. Projeto, No.115, Outubro, pp. 139-142. Serra, G.G. (1989) Post-occupancy evaluation at the urban scale in Brazil, W.F.E. Preiser (ed.), Building Evaluation, New York, Plenum Press, pp.307 -315. Silva, I.A. (1987) A Crise da Moradia. Rio de Janeiro, Livraria Agir Editora. Singer, P. (1985) Habitaçao Popular: 0 que fazer? Arquitetura e Urbanismo, Novembro, pp. 97-99. Souza, R. (1985). Açao temolagica e produçao de habitaç6es. Arquitetura e Urbanismo, Novembro, pp. 100-102. Veja (1991) Prancheta Selvagem. Revista Veja, Ano 24, No. 34, pp. 106-109. References Almeida, R. (1985) A Exigência Politica. Arquitetura e Urbanismo, Novembro, pp. 53-56. Bat/ey, R. (1983) Power Through Burocracy: Urban Political Analysis in Brazil. England, Gower Publishing Company. Carvalho, T.C.C. (1985) As dimensoes da habitaçao. Projeto, No.77, pp. 95-103. Cohab-SP (1978) A Casa Prdpria: Atitudes e opini6es dos moradores do conjunto habitacional "Presidente Castelo Branco"- Carapicuiba. Sao Paulo, Diretoria de Recuperaçâ.o Urbana. Darke, J. and Darke, R. (1988) Affordable Housing: Roles for the State and the Community. J. Friedrichs (ed.), Affordahle Housing and Homeless. Berlin, Walter de Gruyter & Co., pp. 43-58. Guglemi, P. (1985) 0 acesso à moradia. Arquitetura e Urbanismo, Novembro, pp. 77-80. refereed pa pers . 87 Casitas: Gardens of Reclamation: The Creation of Cultural/Social Spaces in the 8arrios of New York City Daniel Winterbottom Department of Landscape Architecture, University of Washington In our increasingly multi-cultural society many voices are not expressed in the public landscape. The design ofpublic space often reflects a class based system. In New York City, cam munity garden spaces are being created on "vacant" land. These spaces serve many community uses including: social, political cultural and religious events; children's play areas; and the raising offood crops and ornamentals. Created by Puerto Rican immigrants, these lush oases, recreations of their indigenous landscapes, provide respite from the surrounding streets. Casitas, investigates how these spaces are used ta express and affirm traditional cultural expression, offering an alternative ta traditional western gardens and public parks. and providing meaningful places for the local communities. garden plots, open spaces, and pathways surrounding a casita de madera or little wooden house. The structure serves as a clubhouse and the garden provides space for ornamental and crop production and socializing. These spaces, created by members of the surrounding Puerto Rican communities serve a variety of cultural, social and political uses. The name casitas derives from the house structure, however the term is now commonly used in New York to describe the entire plot including the "little house" and aU other built structures, ornamental and production gardens, play and social areas and the variety of site elements including shrines and art works. As folklorist, Joseph Sciorra states "The Casita is not merely a nostalgic lament for an idealized past but a form of community organizations whereby control of one's immediate environment is achieved through the use of traditional expressive culture. "1 The building of and attraction ta the casita is rooted in the celebration of traditional Puerto Rican culture and represents a reclamation and recontrol of the environment by the Puerto Rican community. Introduction In the 1980s as residents in the barrios of NYC reclaimed and planted vacant lots, connections to their cultural and natural worlds were deepened, bringing order and wholeness to a deteriorating quality oflife. The sites are commonly referred to as Casitas. They are a distinct combination of vernacular architecture, landscape and art typically composed of 88 . edra 29/1998 El Bohio Boricua, Brooklyn, NY Photo Ejlat Feuer The casita differs from traditional community gardens in its use, in the allocation of space and in the structural patterns. While the primary activity of most co mm unit y gardens is the cultivation of produce and ornamental plantings, the casita serves additionally as classroom, daycare/play area, performance space, ceremonial space and as a community center. AU the casitas on city land must be opened to the public with set hours posted. Of the casitas studied, most are used by extended familymembers, friends and those associated by geographical proximity, and, once established in the community, the user group typically broadens. One casita, "Catano:' is used exclusively by people who come from that town in Puerto Rico. As evolving cultural prototypes, these gardens offer an alternative to traditional western gardens often derivative of European gardens of wealth. These gardens are not simply recreations of vernacular Caribbean gardens but are derivations and adaptations of the traditional forms. As the folklorist Joseph Sciorra, states "the casita is not merely a nostalgic lament for an idealized past, but a form of community organization whereby control of one's immediate environment is achieved through the use of traditional expressive culture." 1 This culture draws from a necessary resourcefulness, tenacity and craft in the use of salvaged materials and plant cultivation to transform a vacant lot into a designed place. The casitas create a social focus, fostering memories and building meaningful connections for a displaced community. They derive from a unique blend of collective and individual expressions. It is this personal quality, woven into traditional typologies, that make these gardens so extraordinary. In this paper l will present findings stemming from a Graham Foundation-funded study. The methodology used includes historical research, in depth sur veys conducted with the principle organizers of the casitas, informaI interviews and bi-lingual questionnaires completed bythe users, field observation, site surveys, and photographie documentation. The historie al material reviewed included the cultural and natural history of Puerto Rico, the evolution of vernacular architecture and associated land uses patterns, the immigrant experience and current conditions of urban Puerto Rican populations. A written survey was developed and administered in the field. The survey method included multiple choice and written response questions. The questions included a short section profiling the users and a longer section focusing on how and why the space is used and valued. The interviews elicited the history of the site and explored at a greater depth the questions in the survey. For each of the casitas field measurements were taken and a site plan was created for comparative interpretation. At least four observation sessions were held at each study site with an average length of three hours, though many visits were longer and more frequent. The observational sessions were held in the spring and fall of 1996 and 1997 in which activities and frequency of use were noted and inventories of plants, structures and spaces were conducted. In this studyeleven casitas in four neighborhoods were documented, including Williamsberg in Brooklyn, NY, The Lower Eastside (Loisiada) and El Barrio known as Spanish Harlem both in Manhattan, and the Melrose section in the South Bronx. The neighborhoods are similar in many respects. Each has a large population of first or second generation Puerto Ricans. Residents income falls significantly below the New York City median and each neighborhood tends to have higher rates of crime and drug abuse. The neighborhoods vary in scale with the tenement housing typology dominating. The quantity of vacant land is greatest in the South Bronx. Of the eleven gardens studied three were initiated by single women, three by a husband and wife team, three by single men and two by groups of men andwomen. The rich heritage of Puerto Rican culture influencing the casitas evolved from the merging of three peoples: the indigenous Taino Indians, the Spanish conquistadors, and the African people brought as slaves to the island bythe Spaniards. The clubhouse casita structures derive from rural housing types that can be traced 500 years to the Taino Indians. The Taino bohio, a two-room rounded structure was made of timbers and sheathed with cane or yagua and roofed with thatch. With Spanish influence, the form became rectilinear and a porch was added. With the concentration of wealth in the cities, the bohio evolved toward housing for the poor. It retained its form principally in the use of certain building materials and the simple technology of construction and in sorne aspects of spatial order. Built from scraps of wood and other found materiaIs, it had an improvised, extemporaneous character. In the 1800's, the Jibaro, people of mixed ancestry practicing subsistence farming, built their bohios refereed papers . 89 on raised posts near rivers to prevent flooding. Unlike the independent Jibaros who had sharecropper rights, Arrabal builders were displaced into urban areas in the 1920s. They claimed land appearing to be vacant, and utilized every available material, constructing a one room house, surreptitiously at night. The strategies ofboth the Jibaro and the Arrabal are similar in the method of assemblage and the accrual mode of construction. While the traditional bohio was still erected in the rural parts o( the island into the 1940s, today they remain only in museums. However, as anthropologist Carol Jopling notes "they remain important in cultural memory not only as evocations of Puerto Rican history, but as significant influences on the spatial order and technology of subsequent vernacular houses."2 The most pronounced African influence was the "changed and increased use of the batey;' an open space fronting the casita. For the Taino, the batey consisted of an open space around which the bohios were grouped and was used as a yard with gardens and fruit trees. As Jopling points out "The term batey itself retained a certain emotional significance, but the land came to function more as a 'commons' or a plaza suitable for social and political activities:'3 Fundamentally, the connection for Puerto Rican culture is a connection to nature, to the rural landscape, to an engagement with the land. This connection can be traced to the religious life of traditional rural communities that are disappearing on the island. While Catholicism prevails among Puerto Ricans, not all practice worship formaIly. The relationship to the church in Puerto Rico is based on individual revelation through reading of the scriptures rather than dependence upon traditional figures of chur ch authority. This was most pronounced in the Jibaros, who developed their own response to Spanish institutions. As Ana Maria Diaz-Stevens describes, this variation of Catholicism drawn from Jibaros religious beliefs centered on "the cycle of nature and committed to an often difficult struggle for subsistence. The people of the rural areas fashioned a religious expression around two things, without which their existence would not be possible-land and family. Thus religious practices and nature's cycles were very closely tied together. The planting and pasturage season, for example usu- 90 . edra 29/1998 ally began in early February and was initiated by the feasts of the Purification of Mary and of St. Blaise."4 This relationship of religion to the land can be seen in both the Santos found in many of the gardens and in the religious fiestas held in the Casitas. In addition Afro-Caribbean beliefs called Santeria and Spiritualism featuring the use ofherbs are practiced in the Casitas. The findings in this investigation concern five qualities of experience significant to the users of the casitas, refuge,recreation, cultivation, celebration and expression. Refuge The casita is an aesthetic, social and spiritual oasis. The garden consists of edible, ornamental and medicinal plantings with opportunistic species, presenting a contrast of wild and cultivated, the whole offering a natural counterpoint to the hardness of the environment. The casita provides an essential place to relax. One women responded "It has changed me in such a way that when 1 feel nervous l would go to the garden and forget everything. l go back home a new person .. :' Another offered "The plants signify something for us, nature helps release stress:' Both being within the garden and the act of cultivating it were regarded as effective for reducing anxiety. This connection between the landscape and human physical and physiological health recalls the work of Roger UIrick,s the Chaplain's and others. 6 In this case, the source of the anxiety is a milieu of poverty, unemployment, substandard hous- El Bohio Boricua, Brooklyn, NY Photo Ejlat Feuer ing, drug trafficking, gang and other criminal activity. While oasis, the casita is also fort. Within the casita,limits on activity and behavior are upheld and enforced. Drugs, fighting and even swearing and other behaviors considered disrespectful are not tolerated. Women respondents spoke of the value of safety concerns primarily for their children. For many mothers, the casita is a place where their children can safely play under other adult's supervision, leaving the mothers time to socialize and recreate. For many of the male users, the primary safety issue was the avoidance of confrontation. The source for this can be found in the traditional social mores and values of Puerto Rico, specifically the concepts of"respectd' and "dignidad:' As defined by social scientist, Anthony Laura, respecto "signifies proper attention to the requisites of the ceremonial order of behavior, and to the moral aspects ofhuman activity. The verb form, repeater, indicates that in anyencounter, one expresses deference to the person whom he confronts:'7 In a study by the Puerto Rican Forum, 'dignidad' "is sometimes explained as the display of a decent regard for the individual regardless of his social position." While one may be poor in Puerto Rico, the association of wealth to respect is less pervasive on the island. For many of the users the transition between cultures is demoralizing and humiliating. The Casitas provide a cultural enclave, a structured environment in which tradition al Puerto Rican aspects of socialization and behavior are accepted. Cultivation For a significant number of the users, the garden is the most valued space within the casita. In answer to the question "Why do you spend time at the casita?", the maj ority responded "to spend time with family and friends:' When asked "what do you do there?" seventy percent of the users answered that gardening was their primary activity. This was substantiated when fifty percent answered that it was the plants or the garden that they most liked in the casita. When asked if any objects had special meaning eighty percent identified the garden as special, particularly the process of planting and the observation of plant growth. Thirty percent responded that the garden made them less nervous, happier, and was "therapeutic:' Surrounded by a high stress environment and with a lack of accessible natural places, the importance of the garden as a stress reducer appears to be important. In both the questionnaires and the interviews, responses such as "The Casita reminds me of my island,I feel as ifI was in Puerto Rico;' or"lts important for our cultural identify because it represents the way our ancestors and grandparents used to live;' or "Yes, it's important because we mustn't forget about our culture and customs" were heard. In an even wider realm, the Casita is a social resource center. In the El Bohio Boricua, Harry Lefond, a community activist helps others by translating and explaining official forms and papers for Medicare, welfare, and housing problems that for many are confusing. The Casitas have also become part of the Latino political network. Community activists and politicians use the gathering spaces to hold forums. Tending the Plots at Humacao, El Barrio, NY Photo Daniel Winterbottom refereed pa pers . 91 We found four kinds of planting in the casitas. The first, food species are planted for consumption and represent 50% to 80% of the cultivated areas. The second group, ornamentals, include perenniais and annuals, shrubs, and shade trees. The third group, herbs and medicinals, are found in aH the gardens. These include herbs for cooking and medicinals, consumed as beverages or used for medicinal baths. Sorne are used by practitioners of Spiritism or Santeria for their healing ceremonies. The fourth group are opportunistic plants at the edges of the sites and a few of these are also used as medicinals. Most species grown are not indigenous to Puerto Rico although a few gardeners acquire seeds from the island. Sorne, not native to Puerto Rico, visually resemble plants found on the island. As Gernaro Lanausse, of Batey Dona Provi notes, the Russian Olive tree in his garden "looks very similar to one of our bean trees. The candules, its the same leaf, same color, same everything. For everybody who passes by that's Puerto Rican, they say 1 know that tree, when is it going to have fruit?" Thus the aesthetics, despite species differences seems to reconnect users with their culturallandscape While the style of gardening is very personalized, sorne general patterns of cultivation are found. Food and herb cultivation is typically done in raised beds and plantings are commonly laid out in rows. AnnuaIs are often mixed into perennial beds. Most shrubs are used as individuals or to define the perimeter of the batey or entry path, but they are rarely massed in groups. This arrangement allows for children's access and reinforces the sense of safety. Most plant selection is limited byaccessibility and affordability. The most common shrubs are roses which are prevalent in Puerto Rico, and gardeners affirm they feel connected to the homeland when the roses bloom. Few of the gardens include lawn. This may be due to availability and high maintenance requirements. Most of the bateys are dirt or partially paved. Planting plots are typically allocated to regular members or to others in the corn munity space permitting. In Neighbors of Vega Baja Garden, plots are set aside for residents of a senior housing complex in the adjacent building. In Batey deI Dona Privy, allocation includes plots for a daycare center, an elementary school, the local beat cop, two 92 . edra 29/1998 neighbors, two relatives and the founder. In this example, horticultural education is provided to the daycare center and elementary school through the casita. This relationship with local schools is not uncommon. In sorne cases, schools provide supplies to the casita in exchange for classes. The casita differs significantly from most city public parks in that it provides open space that residents reclaim and control. The opportunity to partake in the natural and agricultural pro cesses reflects much of the Puerto Rican culture and character. Users feel connected to the place both as a park and "as home:' Participants indicated repeatedlythat their connection to the cycles of planting, cultivating, harvesting and consuming food within the defined space of the garden was very important as a symbol of the Jibaro's subsistence gardens. There was also great pride expressed among older members that their children share a place and experience that they themselves fondly remember, one that represents an indigenous way oflife. Recreation "Play" Anthropologist, Manual Alers-Montalvo, has written "recreation in the Island has been a neighborly, informaI, spontaneous affair. It has consisted in the informaI, unscheduled visit, informaI parties (and) informaI street corner groupS."8 The family unit tends to be extended, close knit and geographically close, so visits and child care among family members and friends is frequent and common. Because of the climate much of the life in Puerto Rico takes place out of doors, in the yards and streets. Daily life in the casita. has the same extemporaneous quality. People stro11 in and out, play dominoes, eat native dishes and listen to tradition al music. In rnany of these neighborhoods, the sidewalk, streets and parks, once used for play, are today perceived to be too dangerous. In sorne casitas play elernents are built or installed. Los Amigos and The Childr-en's Garden were created specifically to provide safe, defined spaces for children's play. These secured play areas relieve sorne of the mothers of their constant supervision, resulting in increased sociability. A nurnber of casitas have special dances for young children celebrating life changes such as birthdays and graduations. These events bring the children together under adult supervision and in a localized community space providing positive alternatives. Celebration For many in the neighborhood, the casita represents the community center where important secular and religious events are celebrated. Religious celebrations are predominantly Christian, but as anthropologist Manuel Alers-Montalvo notes in his study of Puerto Rican Migrants in NY City '1\11 members of a Puerto Rican community-the pious and the indifferent, the Catholic and the Protestant-are constantly exposed to religious activities (processions, festivals, etc.) which although a culturallegacy of Catholicism, have ceased to be Catholic and have become common property of the community at large."9 Many traditionallate winter celebrations oriented to the planting of the fields, are rarely celebrated in the casitas due to weather. Christmas is the exception, and lights, nativity scenes and other decorations are erected in many of the casitas. In Brooklyn, at El Bohio Boricua, the traditional Christmas practice of plena aquinaldos, or gifts in the form of a song, is kept alive. Employing native instruments, the musicians leave the casita and wind their way through the neighborhood stopping at each apartment to sing traditional festive songs and exchange asaltos. In early May, the springtime feast of La Fiesta de la Cruz, and in June, the feast of Saint John the Baptist is celebrated in many casitas. The garden is also the center for a ritual of healing by the community for its members, as exemplified in the leaving of prayers for those in trouble, or the raising of funds to pay for a funeral through the sale of produce. The casita is used to celebrate calendrical holidays induding birthdays, mother's day, graduations, and weddings. Many casitas sponsor softball teams, providing alternatives and mentorships for the youth, and when teams from Puerto Rico visit, fiestas are held in celebration. The ceremony at El Batey de Dona Provi was full of reminiscing through traditional songs and feasting on lechon asado (roast pig) after the Ponce team played the Cayey Tainos from the Bronx. Religious celebrations occur throughout the summer. In Puerto Rico each town has a festival for their Patron Saint and sorne casitas follow this tradition. The transfer of culture from a rural Island to a concrete city requires inventive adaptations. To celebrate the feast of Saint John the Baptist in June, the members of Casita Rojo hold a candlelight procession from the casita to the East River, where participants wash their "bad luck" into the waters, as they do on the beaches of San Juan. Following the deansing, they return to the casita to sing traditional plena songs, dance, and feast. Expression The Santos are Found in Many Gardens in NYC and in Puerto Rico Photo Daniel Winterbottom As a place, every aspect of the casita, induding the architecture, the batey and the garden expresses Puerto Rican culture. Within the casita forms of refereed papers . 93 expression can be found in the visu al art, the music, the dance, and the sociallife, and it is these elements woven together that form the experienced metaphorical sense of place. The arrabal architecture, in its ever evolving manifestations, the gardens, clearly subsistence in form and use, and the animaIs in pens or running loose are all components of a cultural landscape, a "home:' They are also very evocative to those on the "other side" of the fence. Many, unaware of the Puerto Rican memories are lured to the casitas as places of nature. They emit a strong presence transcending cultural barriers and tapping more univers al cultural meanings. Whether recorded salsa, or live bomba or plena, the musical expression of Puerto Rico's black and mestizo working class of the urban arrabales (shantytowns), the rhythmical sounds are a pervasive component of the environment. The participatory nature of the performance stimulates a group expression with the audience picking up improvised materials and complementing the traditional pandetetas and guiros. This group oriented improvisational style is characteristic of the casita, and of the culture. The shared preparation of the food, the raising of the wooden structure, and the shared maintenance of the communal gardens plots are examples. The songs not only bring back memories of the island but, as with the serving of native foods, they create an environment that reinforces the culture, one that is still very much alive, vibrant and familiar, but also one transposed into a new context, and influencing the surrounding environment. In the casita, affirmations can be expressed without threat from the surrounding cultures. This is found not only in the plena music, but in the other creative and material arts that thrive in the casita. At Rincon Criollo in the Bronx, workshops are held in which a variety of hand drums or Panderetas are made. Bomba drums made from recycled pickle and salt cod barrels and Panderetas constructed out of discarded cooking pots or goat skins, imported from Puerto Rico, are produced at the casita. Norma Cruza, traditional dancer, offers formallessons in folkloric bomba and plena dance, and Benny Ayla holds music classes for the local children who often perform in the holiday fiestas. 94 . edra 29/1998 The expressive character of the site itself is most readily apparent in the casitas de madera (Httle wooden house) with its articulated porches, bright pastel house col or and contrasting trim. This colorfuI miniaturized world Îs further expressed in the construction of santos and shrines found in many of the casitas. In the El Jardine lOth Street, a different approach is taken. Salvaged objects are assembled into constructions less specific th an the tradition al santos and wooden houses. The assemblages are very personalized, and like the arrabal architecture are always in process. The combination of painted tires, mannequin parts, doUs and other found objects are juxtaposed into provocative sentinels, spread throughout the site. As sorne have speculated, the process resembles sorne Puerto Rican funerary practices in which toys and common articles are placed on the grave of the deceased, in time creating a collage of pieces that as a whole, tells the story of the pers on. Assemblage in the Rivington Casita on the Lower Eastside, NYC Photo Daniel Winterbottom ',-,other traditional craft practiced at the casita is making of paper mache masks, or vejigante a .\:stumed demon that is worn at carnival festivities :-. Ponce. Many of the casitas, such as the "Children's ~lrden:' stock art making materials for the children. -:-~ey are encouraged to explore their imaginations :i1d to learn from artisans who work with them. -:-l1eir paintings and constructions adorn the casitas ,çith images representative of the island lands cape . :~~,;; .-\s the architecture, garden and music reflect a place :f deep and resonant cultural values, surrounding xick and con crete waUs are often transformed :hrough painted muraIs. Many of these muraIs il.ustrate rural island scenes or views of San Juan and :nany of the casitas are named after the hometown of members or the founder. MuraIs such as the Coqui in the Magic Garden represent symbols of affirmation. The style, less crude and often in the ~and of a local graffiti artist is graphic with symbols, emblems and col ors declaring cultural pride. The muraIs complete the built world of transformation. Conclusion The building of and the attraction to the casita represent the celebration of traditional Puerto Rican culture and the reclamation of the environment. The Taino bohio, the Jibaro subsistence farm, and the Arrabal squatter houses are aU powerful influences on the forms found in the contemporary casitas. And it is the spirit of the independent Jibaro farmer nowa mythic character in the hearts of many Puerto Ricans, and of the Taino, bound to nature for survival that are important cultural influences and affirmations of identity, and that can be se en today in the casitas. The aesthetic quality of the lush gardens, the defined space providing a safe zone and the expression of culture through dance, music and art are aU manifestations of the casita as a refuge. The deep connection to nature stems from an indigenous interpretation of Catholicism, Spiritism, and Santaria. The gardens also serve as educational tools for members of the community, a place to learn principles of cultivation and the medicinal use of plants. Jardine Puertotorriquena, loisaida, (lower Eastside) NY Photo Daniel Winterbottom The casita as a social refuge also serves as a communit y center, created and supported by its members, disseminating information to the community. The garden as a place of cultivation serves a very important role in reducing stress and evokes rich cultural memories. The casita plays an important role as the place of celebration marking major events in the religious and culturallife of the community and its children. The casita is a place reinforcing cultural identity for the "group" but also as a place for individual expression. The Casita as a cultural oasis provides an "alternative to being American in any narrow, culturaUy delimited sense. Beyond its place in pluralist mosaic, Latino culture harbors a deeper cha,llenge to mainstream U.S. society in the form of a different, and potentially divergent, ethos and aesthetic:'lo The Puerto Rican and American flags, typically flown in the casita, symbolize the transference, affirmation and celebration of Puerto Rican culture, and its contribution to the diversity and richness of American culture. It is a living culture, not one oflegend relegated sadlyto a museum, refereed pa pers . 95 but, as the casita illustrates, a culture able to thrive andgrow. The future of the casitas is uncertain. It is not for a lack of will nor purpose that their future may be threatened. But as an island develops a distinct culture, in part due to its isolation, protected from influences from the outside, so too is poverty a form of isolation. As the users of the casitas move up the ladder will assimilation decrease the need for these cultural expressions? How will demands for increased housing effect the use of the vacant lots. In manyrespects these very powerfullandscapes are cultural islands, and their ecologies are very fragile and susceptible to disruption. As they exist, however, they are glorious celebrations of culture, of the natural environment and of an empowered community. References 1. Sciorra, J. (1990) "1 Feel Like l'm in My Country:' Drama Review 34, no(4CT 128). 2. Jopling, C. 1988. Puerto Rican Houses in Sociohistorical Perspective. Knoxville: Tennessee Press. 3. Ibid. 4. Diaz-Stevens A.M. (1996) Aspects of Puerto Rican Religious Experience: A Sociohistorical Overview. In G. Haslip-Viera and S. Baver (Eds.), Latinos in New York: Communities in Transition. Notre Dame: University of Notre Dame. 5. Ulrich, R. (1979) Visual Landscapes and Phycological Well- being.Landscape Research 4(1):17-23. 6. Kaplan S & Kaplan R., (Eds.) (1978) Humanscape: Environments for People. Duxbury Press. 7. Lauria,A. (1964) "Respecto:"'Relago" and Interpersonal Relations in Puerto Rico;' Anthropology Quarterly, 37: 1, 55. 8.Alers-Montalvo, M. (1985) The Puerto Rican Migrants ofNew York City, A Study ofAnomie New York: AMS Press. 9. Ibid. 10. Flores, J. (1996) Puerto Rican and Latino Culture at the Crossroads. In G. Haslip-Viera and S. Baver (Eds.), Latinos in New York: Communities in Transition. Notre Dame: University of Notre Dame. 96 . edra 29/1998