Social Capital Community Benchmark Survey, 2000

DOI

10.17605/OSF.IO/UWMRJ

Citation

Putnam, R. D. (2019, October 29). Social Capital Community Benchmark Survey, 2000.

Summary

The 2000 Social Capital Benchmark Survey was undertaken by the Saguaro Seminar at the John F. Kennedy School of Government at Harvard Unversity. This purpose of the SCCBS, conducted nationally as well as in 41 U.S. communities, is to measure various manifestations of social capital as well as its suspected correlates to (1) provide a rich database for analysis by interested researchers who wish to better understand social capital and (2) provide a tool for communities and organizations to use in program development and evaluation, in part, by enabling relative assessment to other communities and the nation. As a "benchmark" survey, it is the first attempt at widespread systematic measurement of social capital, especially within communities, and it will serve as a point of comparison for future research which attempts to assess changes in key indicators. It is hoped that discussion and use of the survey will also stimulate interest in the broader purpose of fostering civic and social engagement across the country and thus contribute to the revitalization of community institutions.

The survey developed 11 dimensions of social capital for which there are indices including one on religion. The survey also contains 11 measures of religiosity: 1) the respondent's religion (if any) and denomination; 2) how important religion is to the respondent; 3) whether respondent is a member of a religious group; 4) frequency of religious attendance; 5) religious participation, outside of weekly services; 6) level of religious volunteering; 7) level of religious giving; 8) trust of co-congregants; 9) whether respondent has a personal friend of a different religion; 10) degree to which house of worship gives respondent a sense of belonging; 11) whether respondent is an active member in a religious group (other than his/her house of worship). For more information, visit the SCCBS website.

The ARDA has added six additional variables to the original data set to enhance the users' experience on our site.

Data File

Cases: 29233
Variables: 333
Weight Variable: FWEIGHT

Data Collection

July through November, 2000 (except for the West Oakland, California survey which ran from December, 2000 through February, 2001)

Funded By

The national survey was funded jointly by a grant from the Ford Foundation; the community surveys, by 34 local community foundations. Technical support, management, and some initial data analysis was provided by staff at the Saguaro Seminar, John F. Kennedy School of Government - Harvard University.

Collection Procedures

The survey was conducted by telephone using random-digit-dialing (RDD) from July to November, 2000 (except for the West Oakland, California survey which ran from December, 2000 to February, 2001). Interviewing in the national survey and in most of the community surveys was concluded in October. TNS Intersearch, an international survey firm, was commissioned to conduct the interviewing, prepare the data for analysis, and provide a "banner report" (detailed cross-tabulations). Interviews averaged 26 minutes in length.

Sampling Procedures

Each sponsoring organization decided what specific area(s) were to be surveyed, how many interviews to conduct, and if specific areas or ethnic groups were to be over-sampled. In most cases, the survey area was one county or a cluster of contiguous counties; some of the community samples are municipalities and others are entire states. Most of the community surveys used proportionate sampling, that is, no over- or under-sampling of sub-areas or population groups. Most of the samples range in size from 500-1,500 interviews. The national sample (N = 3,003) contains an over-sampling of black and Hispanic respondents to total at least 500 blacks and 500 Hispanics in all. This required screening to identify households with black or Hispanic residents: Several hundred additional blacks (288) and (294) Hispanics had to be identified and interviewed.

Principal Investigators

Robert D. Putnam, Peter and Isabel Malkin Professor of Public Policy at the John F. Kennedy School of Government at Harvard University

Notes

This data set contains a number of variables (scales and indices) that were constructed from other variables in the survey. These notes describe how they were constructed.

1.a. Composite Indices Included in Survey Data Files

In all the indices that were created, effort was made to combine questions and variables that were naturally grouped together in respondent answers. In this sense, respondent answers, rather than preconceived notions, drove the formulation of these indices. In a few cases, like ORGINTER, FAITHBAS, and CIVPART, the original indices were constructed prior to having the full dataset. Analysis of the full dataset proved that there were better variables that captured these dimensions, and those are noted as well.


The following variables were computed from the raw survey responses and included in the SPSS data files (but not shown on the printed survey questionnaire):


Social trust (SOCTRUST) - TRUST TRNEI TRWRK TRREL TRSHOP TRCOP (QUESTIONS 6, 7a, 7b, 7c, 7d, 7f on the original survey covering general interpersonal trust, trust neighbors, trust co-workers, trust fellow congregants, trust store employees where you shop, trust local police). At least three of these answers had to be provided for a score to be calculated. The index is calculated as the mean of the standardized responses to the 5 questions, using national norms to standardize: ITRUST, ITRNEI, ITRWRK, ITRREL, ITRSHOP, ITRCOP. (Note: SOCTRUST is SOCTRST with the polarity of the coding reversed so that higher values on SOCTRUST indicate higher social trust.)


Compute ITRUST = (TRUST2-1.99)/0.96. [Note: Q6 (TRUST) was first recoded into a new variable TRUST2 to reverse the ordering of values: (2=3) (3=2).]

Compute ITRNEI = (TRNEI-1.75)/0.89.
Compute ITRWRK = (TRWRK-1.68)/0.85.
Compute ITRREL = (TRREL-1.38)/0.68.
Compute ITRSHOP = (TRSHOP-2.02)/0.87.
Compute ITRCOP = (TRCOP-1.74)/0.90.

[Note: The subtrahends and divisors in the above formulas are the means and standard deviations from the national social capital survey so that the scores are normed using U.S. parameters.]

Compute SOCTRST = Mean.3(ITRUST,ITRNEI,ITRWRK,ITRREL,ITRSHOP,ITRCOP).
[Note SOCTRUST was then set up to be (-1*SOCTRST).]

Racial trust (RACETRST) - Composite mean trust across 3 of the following 4 groups - Hispanics, Asians, non-Hispanic blacks, and non-Hispanic whites (excluding only respondent's own group) - using Q7g-7j. At least two of these answers had to be provided for a score to be calculated.

RACETRST was created using the following SPSS syntax:
If (ETHNIC4 = 1) RACETRST = Mean.2(TR2BLK, TR2ASN, TR2HIS) .
If (ETHNIC4 = 2) RACETRST = Mean.2(TR2WHT, TR2ASN, TR2HIS) .
If (ETHNIC4 = 3) RACETRST = Mean.2(TR2WHT, TR2BLK, TR2HIS) .
If (ETHNIC4 = 6) RACETRST = Mean.2(TR2WHT, TR2BLK, TR2ASN).

(Note: TR2WHT, TR2BLK, TR2HIS, and TR2ASN represent the recoded versions of TRWHT, TRBLK, TRHIS, and TRASN which reverse the polarity of the scoring so that high values equals higher trust.)

Inter-racial relative trust pairings (RTSTWHT, RTSTBLK, RTSTASN, RTSTHIS)

The variable TRUSTOWN was first computed so that the interracial relative trust pairing variables could be constructed:

If (ETHNIC4=1) TRUSTOWN=TR2WHT.
If (ETHNIC4=2) TRUSTOWN=TR2BLK.
If (ETHNIC4=3) TRUSTOWN=TR2ASN.
If (ETHNIC4=6) TRUSTOWN=TR2HIS.

The relative trust indices (RTSTWHT, RTSTBLK, RTSTASN, RTSTHIS) compare trust of other groups to trust of own group and are calculated as trust of other group minus trust of own group (TRSTOWN). Each respondent gets a score for each of the other three groups she/he rates: trust less, trust more, trust the same.

Example: A black respondent trust whites "some" (code 2 to TR2WHT) and trusts blacks "a lot" (code 3 to TR2BLK). Since that respondent trusts whites less than his own group (trust other - trust own) [2-3=-1], he would receive a "trust less" score on RTSTWHT. Following the formula, negative differences correspond to trusting other group less, differences of 0 signify equal trust, and positive differences represent trusting other group more. Note: All negative values on these measures have been collapsed (recoded) to equal -5, and all positives to equal +5.

Civic participation (CIVPART) - VOTEUS PETITION RALLY PROJECT MARCH - Questions 23, 26a, 26b, 26c, 26d on the original survey covering voting, sign a petition, attend political meeting/rally, work on community project, demonstrate/protest/boycott or march); the index is a simple count of the number of activities engaged in. In SPSS:

Count CIVPART = VOTEUS PETITION RALLY PROJECT MARCH (1).

(Note: Subsequent examination has shown that CIVPART might not be as useful a measure as earlier thought. See additional suggested indices, ELECPOL and PROTEST, below).

Diversity of friendship set (DIVRSITY) - this index is a count of how many different kinds of personal friends the respondent has from the 11 possible types in the Q55 series:

Count DIVRSITY = BBUS BWORKER BWELF BVACH BREL BWHT BHISP BASN BBLK BGAY BLEADER (1).

Number of formal group involvements (GRPINVLV AND GRPINVL2) - counts of "yes" answers in the 18-item Q33 series (GRPINVLV) plus Q32 (GRPINVL2):

Count GRPINVLV = GRPFRAT GRPETH GRPPOL GRPART GRPHOB GRPSELF GRPWWW GRPOTHR GRPREL GRPSPORT GRPYOUTH GRPPTA GRPVET GRPNEI GRPELD GRPSOC GRPLAB GRPPROF (1).

Count GRPINVL2 = GRPFRAT GRPETH GRPPOL GRPART GRPHOB GRPSELF GRPWWW GRPOTHR GRPREL GRPSPORT GRPYOUTH GRPPTA GRPVET GRPNEI GRPELD GRPSOC GRPLAB GRPPROF RELMEM2 (1).

Faith-based social capital (FAITHBAS) - RELMEM RELATEND RELPART1 GRPREL - Questions 30, 31, 32, 33a on the original survey covering church membership, church service attendance, non-religious service church participation, affiliation with non-church religious groups). At least two of these answers had to be provided for a score to be calculated. The index is calculated as the mean of the standardized responses to the 4 questions, based on national norms.

Note: NORELSVC is a recode of Q31 (RELATEND) that translates reported religious service attendance into estimated days per year ("Every week" = 52, "Almost every week" = 39, "One or twice a month" = 18, "A few times a year" = 3, "Less often than that" = 0). To construct FAITHBAS, additional new variables were first created and included in the data files so that persons answering "no religion" (RELIG Q29) would be considered as being a church NON-member, attending religious services less often than a few times a year, and not having taken part in any other church-related activity. Along with the standardized version of GRPREL Q33a, the standardized versions of these variables were used in the computation of FAITHBAS: IRELMEM2, IRELATTE, IRELPART, and IGRPREL:

Compute IRELMEM2 = (RELMEM2-1.42)/0.49.
Compute IRELATTE = (NORELSVC-28.99)/21.81.
Compute IRELPART = (RELPART2-1.60)/0.49.
Compute IGRPREL = (GRPREL-1.84)/0.37.
[The subtrahends and divisors in the above formulas are the means and standard deviations from the national social capital survey so that the scores are normed using U.S. parameters.]

Compute FAITHBAS = Mean.2(IRELMEM2,IRELATTE,IRELPART,IGRPREL).
Note: An improved, slightly more coherent version of faith-based social capital results from adding standardized measures of religious organization based contributing (ZGIVEREL) and volunteering (ZVOLREL2):

Compute FAITHBA2 = Mean.3(IRELATTE,IRELMEM2,IRELPART,IGRPREL, ZGIVEREL,ZVOLREL2).
Organized group interactions (ORGINTER) - (CPUBMEET CCLUBMET CPARADE - Questions 56l, 56e, 56a on the original survey covering attending public meetings, attending club meetings, attending local community events). At least two of these answers had to be provided for a score to be calculated. The index is calculated as the mean of the standardized scores of the 3 questions, based on national norms: IPUBMEET, ICLUBS, IPARADE.

Compute IPARADE = (PARADE2-7.25)/12.26.
Compute ICLUBS = (CLUBS2-6.04)/12.46.
Compute IPUBMEET = (PUBMEET2-2.55)/6.16.
[The subtrahends and divisors in the above formulas are the means and standard deviations from the national social capital survey so that the scores are normed using U.S. parameters.]

Compute ORGINTER = Mean.2(IPARADE,ICLUBS,IPUBMEET).

Note: MACHER, described in the next section, is a more coherent alternative to ORGINTER and thus recommended.

Informal social interactions (SCHMOOZ) - (CFRDVIST CFAMVISI CJOBSOC CFRDHANG CCARDS - Questions 56f, 56d, 56h, 56i, and 56c on the original survey covering having friends visit home, visiting with relatives, socializing with co-workers outside of work, hanging out with friends in public places, playing cards and board games). At least two of these questions had to be answered for a score to be calculated. The index is calculated as the mean of the standardized responses to the 5 questions, based on national survey norms: IFRNDHOM, IFAMVIS, IJOBSOC, IFRNDHNG, ICARDS.

Compute ICARDS = (CARDS2-11.67)/17.91.
Compute IFAMVIS = (FAMVIS2-25.02)/22.38.
Compute IFRNDHOM = (FRNDHOM2-22.10)/21.40.
Compute IJOBSOC = (JOBSOC2-14.08)/18.42.
Compute IFRNDHNG = (FRNDHNG2-15.51)/19.60.
[The subtrahends and divisors in the above formulas are the means and standard deviations from the national social capital survey so that the scores are normed using U.S. parameters.]

Compute SCHMOOZ = Mean.2 (ICARDS,IFAMVIS,IFRNDHOM,IJOBSOC,IFRNDHNG).
Each of the above interval-level indices were also divided into ranges (3 or 4 categories such as high, medium, low) for use in categorical tables/analyses. These variables' names are:

Social trust STRSTCAT
Composite racial group trust RACETCAT
Civic participation CIVPCAT
Diversity of friendship network DIVRCAT
Group involvements (without church participation) GRPINCAT
Group involvements (with church participation) GRP2CAT
Faith-based social capital FAITHCAT
Organized group interactions ORGINCAT
Informal social interaction SCHMZCAT

1.b. Other Social Capital Measures Not Included in Data Files

Several additional measures of social capital were computed and found to be useful. These were constructed after the initial data files had been prepared and sent but are contained in the updated file "combdat2":

Giving and Volunteering -- This index combines reversed polarity versions of volunteering for different types of organizations: arts (VOLARTS Q59e), health-related (VOLHEA Q59b), neighborhood (VOLNEI Q59f), religious (VOLREL Q59a), youth groups (VOLYOU Q59c), and those which help the poor or elderly (VOLHUM Q59d), the total number of times volunteered (VOLTIMES Q58), and contributions to secular charities (GIVEOTHR Q37b) and religious causes (GIVEREL Q37a). At least 5 responses had to be given for a score on CHARITY to be computed. The following SPSS syntax statement was used:

Compute CHARITY=9*(Mean.5(VOLARTS2,VOLHEA2,VOLHUM2,VOLNEI2, VOLREL2,VOLYOU2,(VOLTIME2/12),GIVEOTHR,GIVEREL)).
Note: in the current data file, there are two versions of CHARITY that you might want to use: CHARITY2 is the continuous variable whose equation is given above. CHARITY3 breaks CHARITY2 into Low, Medium, and High (and is thus the analog for CHARITY of, for example of what DIVRCAT is to DIVRSITY).

Electoral Politics -- This index combines past voting (VOTEUS Q23), voter registration (REGVOTE Q22), interest in politics and national affairs (POLINT Q21), political knowledge (of U.S. Senators) (POLKNOW Q28), and frequency of newspaper reading (PAPER Q17). In SPSS syntax:

Compute ELECPOL=5*(Mean.3(VOTEUS,REGVOTE,((POLINT-1)/3),
((POLKNOW2-1)/4),(PAPER/7))).

Note: in the current data file, there are two versions of ELECPOL that you might want to use: ELECPOL2 is the continuous variable whose equation is given above. ELECPOL3 breaks up ELECPOL2 into Low, Medium, and High (and is thus the analog for ELECPOL of, for example of what DIVRCAT is to DIVRSITY).

Protest - This index is a measure of non-electoral forms of political participation: signing petitions (PETITION Q26a), attending political meetings or rallies (RALLY 26b), joining in any demonstrations, protests, boycotts, or marches (MARCH 26d); also, involvement in local reform efforts (REFORM Q34), membership in political groups (GRPPOL 33m), ethnic, nationality, or civil rights groups (GRPETH 33l), or labor unions (GRPLAB 33i). In SPSS syntax:

Compute PROTEST=7*(Mean.4(MARCH,GRPPOL,RALLY,REFORM,
GRPETH, PETITION,GRPLAB)).

Note: in the current data file, there are two versions of PROTEST that you might want to use: PROTEST is the continuous variable whose equation is given above. PROTEST3 breaks up PROTEST into Low, Medium, and High (and is thus the analog for PROTEST of, for example of what DIVRCAT is to DIVRSITY).

Macher - This index consists of the factor score resulting from a principal components analysis of GRPINVLV, OFFICER, CLUBS2, and PUBMEET2.

Note: in the current data file, there are two versions of MACHER that you might want to use: MACHER is the continuous variable whose equation is given above. MACHER3 breaks up MACHER into Low, Medium, and High (and is thus the analog for MACHER of, for example of what DIVRCAT is to DIVRSITY).

1.c. Other Constructed Variables in Data Files

Apart from the social capital indices (and related variables) described in the previous section, several other variables which do not appear on the questionnaire were computed or appended and included in the data files:

REFCON - Designates if the interview resulted from a refusal conversion.
EDUC_ALL - Combines EDUC Q61 (education level) and EDUCA Q61A (if received GED or high school equivalency) into one education measure which applies to all.

RACE_ALL - Combines RACE Q63 (race of non-Hispanics) and HISPRACE Q62B (race of Hispanics) into one race measure which applies to all respondents.

METSTAT - A "metropolitan status" (MS Area) code provided for each sample telephone number, measuring location of place of residence relative to MSA center city or if not part of an MSA: center city of MSA, same county as MSA center city but not center city, other county of MSA, in MSA with no center city, and non-MSA.

ETHNIC4 - A recoding of race and Hispanic ethnicity into four primary racial/ethnic categories: non-Hispanic white, non-Hispanic black, Hispanic, and Asian.

POLKNOW2 - A recoding of POLKNOW Q28 responses (POLKNOW) to form an ordinal "political knowledge" scale.

AGE - A recoding of BYEAR Q60 (year of birth) into age.
AGERANGE - A collapsing of BYEAR Q60 (year of birth) into 4 age ranges.
COMMUNTY - Code representing the community sampled or national sample.
FWEIGHT - Final weight applied to the case/respondent.
RESPNO - Respondent's unique identification number.
CALLD - Date of interview.
SCREEN - Indicates whether interview occurred as a result of screening for ethnic minority status or specific geographic location.

SAMP - Code representing the community stratum sampled.
CENSDIV - Code representing one of nine U.S. Census divisions.
CENSREG - Code representing one of four U.S. Census regions.
CALL - Number of contact attempts required to complete interview.
LANGASK - Language of interview (English or Spanish).
CITY - Respondent's report of city of residence (applies in Delaware and Greensboro samples only).

CITY2 - Updated classification of Wilmington residence (applies in Delaware sample only).
Frequency of Civic/Social Activities Which Combine Follow-Up Probe
PARADE2 - Q56A: CPARADE and PARADE.
ARTIST2 - Q56B: CARTIST and ARTIST.
CARDS2 - Q56C: CCARDS and CARDS.
FAMVIS2 - Q56D: CFAMVISI and FAMVISIT.
CLUBS2 - Q56E: CCLUBMET and CLUBMEET.
FRNDHOM2 - Q56F: CFRDVIST and FRDVISIT.
FRNDRAC2 - Q56G: CFRDRAC and FRDRAC.
JOBSOC2 - Q56H: CJOBSOC and JOBSOC.
FRNDHNG2 - Q56I: CFRDHANG and FRDHANG.
TEAMSPT2 - Q56J: CSOCSPRT and SOCSPORT.
WWWCHAT2 - Q56K: CWWWCHAT and WWWCHAT.
PUBMEET2 - Q56L: CPUBMEET and PUBMEET.
LOCVIS2 - Q57: CLOCVIS and LOCVIS.
VOLTIME2 - Q58: CVOLTIME and VOLTIMES.

Interviewers asked how often the respondent engaged in each activity (PARADE ARTIST CARDS FAMVISIT CLUBMEET FRDVISIT FRDRAC JOBSOC FRDHANG SOCSPORT WWWCHAT PUBMEET LOCVIS VOLTIMES Q56A-L, 57, 58) in two different ways. First: How many times in the past 12 months have you [participated in activity X]? Those who could not, or did not, answer the initial, open-ended question were asked a follow-up, probe specifying frequencies to try to make it easier for the respondent to provide an answer: Would you say you never did this, did it once, a few times, about once a month on average, twice a month, about once a week on average, or more often that that? For each item, the variable name for the initial question ("How many times have you...") begins with a "C" (e.g., CPARADE), the variable name for the probe is similar but omits the "C" (e.g., PARADE), and the variable which combines the two distributions (those answering the initial question and those answering the probe) ends with a "2" (e.g., PARADE2). Users will most commonly want to use the combined variables in their analysis. [Note: for the combined variables in this series (of the form PARADE2, ARTIST2, etc.) answers of "53" to the actual numerical estimate (e.g., CPARADE) and answers of 9 ("more than every week") to the range-prompted questions (e.g., PARADE) were coded as 60 times per year.]