All the Movements Fit to Print: Who, What, When, Where ...oliver/SOC924/Articles/AmentaEtAl2009A… · the SMO family level, coverage correlates highly with common measures of the

Gaining the mass news mediarsquos attention iscritical to the struggles of political advo-

cacy and social movement organizations(SMOs)1 gaining coverage is a measure of an

SMOrsquos cultural influence (Berry 1999 Ferreeet al 2002 Gamson 2004 Gamson et al 1992Gitlin 1978 1980 Koopmans 2004 Lipsky1968 Vliegenthart Oegma and Klandermans

All the Movements Fit to Print Who What When Where and Why SMO Families Appeared in the New York Times in the Twentieth Century

Edwin Amenta Neal CarenUniversity of California-Irvine University of North Carolina-Chapel Hill

Sheera Joy Olasky James E StobaughNew York University University of California-Irvine

Why did some social movement organization (SMO) families receive extensive media

coverage In this article we elaborate and appraise four core arguments in the literature

on movements and their consequences disruption resource mobilization political

partisanship and whether a movement benefits from an enforced policy Our fuzzy-set

qualitative comparative analyses (fsQCA) draw on new unique data from the New York

Times across the twentieth century on more than 1200 SMOs and 34 SMO families At

the SMO family level coverage correlates highly with common measures of the size and

disruptive activity of movements with the labor and African American civil rights

movements receiving the most coverage Addressing why some movement families

experienced daily coverage fsQCA indicates that disruption resource mobilization and

an enforced policy are jointly sufficient partisanship the standard form of ldquopolitical

opportunityrdquo is not part of the solution Our results support the main perspectives while

also suggesting that movement scholars may need to reexamine their ideas of favorable

political contexts

AMERICAN SOCIOLOGICAL REVIEW 2009 VOL 74 (August636ndash656)

Direct correspondence to Edwin Amenta Departmentof Sociology University of California 3151 SocialScience Plaza Irvine CA 92697-5100 (ea3uciedu)We thank Rogers Brubaker Bobby Chen ElizabethChiarello Nitsan Chorev Stephanie Joy Dialto ThomasElliott Jeff Goodwin Drew Halfmann FrancescaPolletta Kelsy Kretschmer Rory McVeigh MichaelMann Andrew Martin David S Meyer Natasha MiricZiad Munson Charles Ragin Kelly Ramsey WilliamG Roy Rens Vliegenthart Roger Waldinger OwenWhooley and the members of the NYU Department ofSociology PPP Workshop the UCI Social Justice andSocial Movement Workshop the UCLA Comparativeand Historical Workshop the ASR editors and seven

anonymous reviewers for comments on previous ver-sions We thank J Craig Jenkins Debra C Minkoff andJudith Stepan-Norris for sharing data This researchwas supported in part by National Science FoundationGrant SES-0752571

1 Our conceptualization of SMOs includes nation-al advocacy organizations that make claims on or onbehalf of mass constituencies similar to definitionsused by McCarthy and Zald (1977) and Gamson(1990) and scholars following their work We use theterm ldquosocial movement organizationrdquo for simplicityrsquossake although we are cognizant of the fact that otherscholars (notably McAdam 1982) reserve the term fororganizations that threaten or engage in disruptive col-

Delivered by Ingenta to University of Wisconsin-Madison

Wed 28 Oct 2009 035926

2005 review in Earl 2004) In this study weaddress why some SMO families receive exten-sive newspaper coverage by developing newdata on New York Times articles that mentionedUS SMOs across the twentieth century Wefirst identify which US SMOs and SMO fam-ilies have received the greatest newspaper cov-erage we then use this information tosystematically address why some movementfamilies receive extensive coverage appraisingwell-known theories of social movements andmovement consequences

Explaining newspapersrsquo SMO coverage isimportant for several reasons SMOs seek topromote many sorts of social change from cre-ating interests and identities to spurring politi-cal participation and civic engagement towinning political goals (Amenta 2006 Andrews2004 Clemens 1997 Gamson 1990 Ganz2000 McAdam 1982 McCarthy and Zald 1977Polletta 2002 Sampson et al 2005 Skocpol2003) and media coverage is important to theseefforts Coverage also constitutes key data inmapping political interests and identities amongthe politically disadvantaged it provides a meas-ure of discursive presence or influence in theproduction of culture akin to Gamsonrsquos (19901998) ldquoacceptancerdquo (Earl 2004) Using fuzzy setqualitative comparative analyses (fsQCA) across2153 movement family years we explore whysome movement families received extensivecoverage employing arguments from the dis-ruption perspective and the resource mobiliza-tion and political contextual theories We alsodevelop a relatively new political contextualargument enforced policies for a movementrsquosconstituency will spur movements and theircoverage

MOTIVATION PREVIOUS WORKAND MODELS OF MOVEMENTINFLUENCE

COVERAGE AS A CULTURAL CONSEQUENCE

OF MOVEMENTS

SMOs have been central to movement researchsince the early 1970s (Gamson 1990 McCarthyand Zald 1977) but few studies go beyondexamining one movement (cf Gamson 1990

Skocpol 2003) Moreover the mass news mediahave the widest gallery of all forums in the pol-icy-making process (Gamson 2004) so theattention SMOs receive in the mass media bol-sters their positions as representatives for theinterests and constituencies they claim (Ferreeet al 2002 Koopmans 2004) The mass mediahelp legitimize SMOs in a democratic politicalsystem in which most organized groups cangain some access to political institutions mediacoverage itself is a demonstration of SMOsrsquoimpact or acceptance (cf Gamson 1990) Manyalso see mass media coverage as necessary formovements to be influential (Lipsky 1968)SMOs seek to showcase and transmit their caus-es to relevant third parties and bystanders(Gamson 2004) by offering alternative framingsof issues (Cress and Snow 2000 Ferree et al2002 Ryan 1991) or discrediting opponentsand their framings (Gamson 2004) SMOs cangain coverage and influence policy debates inmultiple ways aside from protest (Amenta 2006Amenta Caren and Olasky 2005 Andrews2004) and those that receive coverage also tendto gain support (Vliegenthart et al 2005) Inshort media coverage of SMOs across move-ments and over time is an important if limitedconsequence of movements

SMOs appear in newspapers in differentways but always as a function of the practicesof newsgathering organizations which are con-cerned with generating ldquostoriesrdquo and ldquonewsrdquo(see Schudson 2002) Unlike with protest events(see review in Earl et al 2004) there is no wayto compare coverage of SMOs with all theirrelevant activity or all dimensions of their sizeIt is possible however to compare SMO cov-erage with important measures of movementsize such as membership and organizationaldensity and with protest events and other dis-ruptive activities Most important by compar-ing across all SMO families over a century wecan test theories about social movements andmovement consequences to explore why someSMO families achieved high coverage

FOUR THEORETICAL APPROACHES TO

EXPLAINING MOVEMENTS AND OUTCOMES

Prominent ideas in the literature on the conse-quences of social movements suggest first thatdisruption brings influence for movements Inthe classic view (Piven and Cloward 1977)

WHY DID SMOS APPEAR IN THE TIMESmdashndash637

lective action a set of organizations subsumed by ourdefinition


Wed 28 Oct 2009 035926

mass turmoil is expected to influence politicalleaders by creating a threat to the social orderThis point of view dovetails with the literatureon newspaper coverage Newspapers are morelikely to report on large and violent events (Earlet al 2004 McCarthy McPhail and Smith1996 Myers and Caniglia 2004 Oliver andMyers 1999) so organizations linked to dis-ruptive action will likely receive more extensivecoverage (see also Corbett 1998 Rohlinger2002)

The resource mobilization theory (McCarthyand Zald 1977 Zald and McCarthy 2002)expects movements with many organizationsand capacities to be the best mobilized and toexert influence of many different sorts includ-ing media related SMOs and SMO familieswith the most extensive resources would thus beexpected to receive extensive coverage (see alsoCorbett 1998) Newspapers tend to view theirreporting as reflecting main tendencies in socialtrends (Gans 1979) so coverage may be deter-mined in part by the size of SMOs and SMOfamilies Studies of newspaper coverage of col-lective action events indicate that coveragefocuses on events that draw the participation oflarge organizations (Earl et al 2004) Researchidentifies many different aspects of movementsas resources to appraise this approach includ-ing membership in SMOs and SMO families(Zald and McCarthy 2002) particularly thenumber of SMOs in the family available to becovered (Minkoff 2002) From this perspectivethe expectation is that the more members andthe greater the number of organizations avail-able for coverage in an SMO family the greaterthe coverage

Along with these two theories our researchhere also addresses two political contextualmodels that seek to explain movements andtheir consequences The most prominent argu-ment in the literature on political contexts orldquoopportunitiesrdquo expects movements to expandand gain influence with a sympathetic regimein power (Meyer and Minkoff 2004) This is typ-ically understood and modeled in the US con-text as a Democratic regime for movements ofthe left and a Republican regime for movementsof the right In this view ideologically similarregimes should both stimulate movements andpromote consequences favorable to them

An additional although less prominent argu-ment from the political context perspective is

that movements will advance in the wake ofmajor policy changes favoring the movementrsquosconstituency (Amenta and Young 1999 Berry1999 see also Baumgartner and Jones 1993Halfmann Rude and Ebert 2005) In this viewmovements are sustained politically throughpolicies related to their constituenciesMovements are shaped by the rhythms of statebuilding (Skocpol 2003 Tilly 2005) and poli-cy making (Baumgartner and Jones 1993)which alter politics and often work in a self-rein-forcing way (Pierson 2000) These policiesshould bolster movements and help promotefurther outcomes favorable to them

CONCEPTUALIZATIONS DATA ANDMETHODS

We examined the coverage of all national USSMOs in articles in the New York Times fol-lowing a longstanding practice in newspaperstudies of movements (see Earl et al 2004) todetermine which SMOs and SMO families havebeen most publicly prominent in every year ofthe twentieth century Many prominent longi-tudinal studies of movements are based on news-paper data on protest events and use the NewYork Times with its national focus as a source(Jenkins and Eckert 1986 Kerbo and Shaffer1992 McAdam and Su 2002 Soule and Earl2005)

Working from def initions of SMOs byMcCarthy and Zald (1977) and Gamson (1990)our first step was to identify the population ofnational political SMOs contending in the twen-tieth centurymdashno easy task as until now noone has done so (cf Brulle et al [2007] onenvironmental organizations) We then searchedthe New York Times using ProQuest HistoricalNewspapers for mentions of these SMOs inarticles Next we arrayed the data listing organ-izations according to their overall mentionsWe checked the results with data from theWashington Post We then categorized the organ-izations into different groupings based on move-ment type From there we compared measuresof SMO coverage in the Times with other meas-ures of movement size and activity to see howclosely they corresponded to and correlatedwith coverage figures Finally we used fsQCAanalyses to ascertain why some movement fam-ilies received extensive coverage employingfour theories of movement outcomes

638mdashndashAMERICAN SOCIOLOGICAL REVIEW


Wed 28 Oct 2009 035926

To conceptualize SMOs we rely on defini-tions by McCarthy and Zald (1977) and Gamson(1990) who refer respectively to ldquosocial move-ment organizationsrdquo and ldquochallenging groupsrdquoFor McCarthy and Zald SMOs are formalorganizations whose goals are allied with thoseof a social movement For Gamson (see alsoBerry 1999) a challenging group is a formalorganization that seeks to mobilize an unmobi-lized constituency and has an antagonist inauthority outside its constituency These large-ly similar definitions include only politicallyinflected organizations like Gamson we rely onorganizations with national goals These defi-nitions also include most of what today arecalled political advocacy organizations Forinstance Gamsonrsquos large sample netted suchinstitutional-tactic-reliant organizations as theAmerican Association of University Professorsthe Proportional Representation League andthe League of American Wheelmen Andrewsand Edwardsrsquos (2004) ldquoadvocacy organizationsrdquoare similar to the McCarthy and ZaldGamsonversion of SMOs but they also include ldquointer-est groupsrdquo (Granados and Knoke 2005) Wealso include what McCarthy and Zald refer toas ldquoestablishedrdquo SMOs or mobilized challeng-ing groups That is we do not stop includingorganizations such as the AFL-CIO theNAACP NOW and the Sierra Club once theyhave mobilized a new constituency

Needless to say this definition excludes manyorganizations The McCarthy and ZaldGamsondefinition of SMOs we employ does not includeall voluntary mass organizations as do studiesof civic engagement (Putnam 2000 Schoferand Fourcade-Gourinchas 2001 Skocpol 2003)We do not include standard interest groupssuch as Chambers of Commerce think tanksand professional associations SMOs thatengage in or threaten non-institutional or trans-gressive action (McAdam 1982 McAdamTarrow and Tilly 2001) form a distinct subsetour results do not generalize to this subset Wealso exclude the main political parties Unlikein Europe US SMOs in the twentieth centuryhave not ldquograduatedrdquo to become significantnational political parties and they are not main-ly concerned with nominating and electing can-didates to political offices There are many otherways to conceptualize movements and organi-zations (see Snow Soule and Kriesi 2004) butwe chose this definition because of its wide-

spread currency and because these organiza-tions are the most directly influential in insti-tutional politics and elite debates

We started with previous large lists of SMOs(Fountain 2006 Tilly Nd) work that compareslarge numbers of organizations (eg Gamson1990 Minkoff 1997 Skocpol 2003 Snow et al2004 Wilson 1973) many articles and morethan 100 monographs on movements advicefrom colleagues and the Encyclopedia ofAssociations We also inspected newspaper arti-cles with the words ldquogroupsrdquo and ldquoorganiza-tionsrdquo in the headline to identify furthercandidates for inclusion We then searched forall articles mentioning the SMOs throughProQuest using the official name of the organ-ization and its acronyms We examined some ofthe articles indicated and expanded or restrict-ed the search terms for the most accurate countWe cross-checked the Timesrsquo coverage againstcoverage in the Washington Post for each ofthe top 30 SMOs in the Timesrsquo coverage over-all as well as the top 25 SMOs in the Timesrsquocoverage for a given year (see below) All fourauthors coded led by the senior scholars of theteam and pairwise reliability scores were alwaysabove 90 percent2

We identified 1247 qualifying SMOs in thetwentieth century although only 947 had cov-erage in the Times Altogether we identified298359 article mentions of SMOs It may notever be possible to identify all qualifying SMOsbut our search methods make us confident thatwe located almost all qualifying SMOs thatreceived significant national newspaper cover-


2 Some scholars use the IRSrsquos list of tax-exemptorganizations (notably Brulle et al 2007) which inDecember 2006 numbered 677043 We took a ran-dom sampling of 100 organizations from this listand searched for them online locating 80 Of theseonly the Bowhunting Preservation Alliance wasfound barely to meet our criteria for an SMO butappeared in no articles To ensure we captured thecoverage of federated organizations we oftensearched for shortened versions of official namessuch as ldquowomanrsquos suffrage associationrdquo for theNational American Womanrsquos Suffrage AssociationWe also searched for alternatives such as ldquowoman suf-frage associationrdquo and ldquowomenrsquos suffrage associa-tionrdquo We counted any mention of a lower-levelorganization as part of the coverage of the nationalorganization (cf Brulle et al 2007)


Wed 28 Oct 2009 035926

age We are also confident that the potentialfuture identification of SMOs as yet uncoveredwill not greatly change the results below Weemploy individual mentions (cf Vliegenthart etal 2005) for simplicityrsquos sake and also becausethere was little variation among the most cov-ered SMOs in the degree to which they appearedin front-page articles

WHICH SMOS AND MOVEMENTSRECEIVED THE MOST COVERAGE

Which SMOs and movements received thegreatest coverage The SMO with the most cov-erage overall is unsurprisingly the AFL-CIO(including coverage of the AFL and CIO indi-vidually before they merged in 1953) The extentof its dominance is surprising however as itreceives more than three times as many men-

tions as the next SMO the American Legion(see Table 1) The National Association for theAdvancement of Colored People (NAACP) is aclose third and the American Civil LibertiesUnion and the Ku Klux Klan (KKK) round outthe top five each appearing in more than 8000articles The top-30 list also includes sevenother labor-union organizations Other well-known social movements are well representedin the top 30 including four additional SMOsrelating to African American civil rights theNational Council of Churches the NationalUrban League the Black Panther Party and theCongress of Racial Equality Two additionalveterans organizationsmdashthe Grand Army of theRepublic and the Veterans of Foreign Warsmdashrank in the top 30 as well Other movementfamilies are represented by longstanding organ-izations including the feminist (League of


Table 1 Top 30 SMOs with the Most New York Times Coverage in the Twentieth Century withCoverage from the Washington Post

Rank Organization (Year of Founding) Times Front Page Post

01 American Federation of LaborndashCongress of Industrial Organizations 41718 6848 33690mdash(1886 1937 1955)

02 American Legion (1919) 12650 1441 926203 National Association for the Advancement of Colored People (1909) 12616 1707 1224704 American Civil Liberties Union (1920) 8911 1022 743105 Ku Klux Klan (1867) 8067 1119 587906 United Mine Workers (1890) 7044 1397 656007 League of Women Voters (1920) 6869 461 764708 International Ladies Garment Workers (1900) 5875 675 60109 International Brotherhood of Teamsters (1903) 5216 1848 886410 Veterans of Foreign Wars (1936) 4829 480 641911 National Education Association (1857) 4725 462 461612 Anti-Saloon League (1893) 4581 851 253313 United Steelworkers (1942) 4019 392 177714 American Jewish Congress (1918) 3849 297 87615 Grand Army of the Republic (1866) 3492 149 285316 Black Panther Party (1966) 3460 394 233317 American Jewish Committee (1906) 3317 263 107418 Actorsrsquo Equity Association (1913) 3229 157 21619 American Society for the Prevention of Cruelty to Animals (1866) 3016 51 21320 United Auto Workers (1935) 2872 195 525721 National Council of Churches (1950) 2649 256 191922 Anti-Defamation League (1913) 2618 247 142423 Planned Parenthood (1923) 2610 204 219924 International Brotherhood of Electrical Workers (1891) 2541 337 78125 Sierra Club (1892) 2497 218 282226 National Urban League (1910) 2495 300 120327 Congress of Racial Equality (1942) 2349 519 54028 American Federation of Teachers (1916) 2267 325 106329 International Typographical Union (1852) 2130 165 118030 Americans for Democratic Action (1947) 2052 298 2076


Wed 28 Oct 2009 035926

Women Voters) anti-alcohol (the Anti-SaloonLeague) animal protectionrights (AmericanSociety for the Prevention of Cruelty toAnimals) environmental (Sierra Club) andreproductive rights movements (PlannedParenthood)

We also examine the coverage of the topSMOs in the Washington Post Aside from men-tions of SMOs in the Post being lower overallthere are a few important differences A fewNew York-based organizations are far bettercovered in the Times including the AmericanJewish Congress and the American JewishCommittee the Actors Equity Association withits connections to Broadway receives a lot ofattention in the Times but little in the Post Allthe same the correlation between the top-30 listsis 96 with most of the slippage due to the NewYork-based organizations3 Among the top 30moreover the correlation between overall cov-erage and appearing in front-page articles inthe Times is extremely high (97)

From here we analyze coverage according tobroad categories families or industries of socialmovements to ascertain which received the mostcoverage across the century Lacking scholarlyconsensus in both the categories of socialmovements and allocating SMOs to them weemploy frequently used if somewhat broadmovement familiesmdashincluding labor AfricanAmerican civil rights environmental conser-vation and ecology veterans andfeministwomenrsquos rightsmdashfor a total of 34 mutu-ally exclusive and exhaustive categories Due tothe lack of consensus and the small numbers ofarticle counts for some possible movement fam-ilies three of these categories have a residualquality We categorized SMOs that were large-

ly left- or right-wing in orientation but that didnot fit neatly into a more coherent movementfamily as ldquoprogressive otherrdquo and ldquoconserva-tive otherrdquo SMOs seeking civil rights for spe-cific groups but that did not receive enoughcoverage to warrant an entire category are cat-egorized as ldquocivil rights otherrdquo We also focuson issues rather than movementsrsquo demograph-ic makeups organizations largely or exclusive-ly consisting of women might find themselvesas part of the feminist anti-alcohol or chil-drenrsquos rights movements for instance andorganizations of students might be part of anti-war civil rights conservative or progressiveSMO families

Table 2 lists each movement family or indus-try according to the mentions received by theorganizations constituting the category Laborreceived by far the most mentions accountingfor 363 percent of articles in which SMOs werementioned more than three times as much as itsclosest competitor the African American civilrights movement which had 98 percent Laborremains first easily even when individual unionsare not counted with about 189 percent of thecoverage (We also list the movements withoutindividual unions because these organizationsso dominate coverage) Behind these two arefour SMO families the veteransfeministwomenrsquos rights nativistsupremacistand environmental conservation and ecologySMOs each had between 40 and 76 percent ofthe coverage Jewish civil rights civil libertiesanti-war and residual conservative SMOs roundout the top 10 Although the veterans and nativistmovement families place in the top five and theJewish civil rights and civil liberties familiesplace in the top 10 none have received exten-sive scholarly attention

Next we examine the overall trajectory of thetop movement families or industries Figure 1shows the coverage for the labor AfricanAmerican civil rights and veterans SMO fam-ilies (in three-year moving averages to smoothout arbitrary year-to-year variations) For rea-sons of scale we include the labor movementwithout individual unions although the patternis similar (results not shown) Labor has a strongnewspaper presence throughout the centurytaking off in the 1930s and 1940s and declin-ing in the 1950s and beyond although remain-ing at a significantly high level of coverageCoverage of the African American civil rights


3 In the early decades of the century the Post cov-ered veterans more extensively possibly becausethese organizations focused their attention on thecapital although the coverage often relates to law-makersrsquo affiliations The Post is available via theProQuest Historical database only through 1992 sowe use Lexis-Nexis which is available from 1977 for1993 through 1999 For the top-10 covered SMOs ofthe 1980s coverage figures produced by ProQuestand Lexis-Nexis searches are correlated highly at99 and the number of articles is similar withProQuest unearthing 13694 articles and Lexis-Nexis13618


Wed 28 Oct 2009 035926

movement takes off in the 1960s after makinggains in the late 1950s and does not decline untilthe mid-1970s If social movements have movedin waves (Tarrow 1994) labor was at the cen-ter of the wave in the 1930s and 1940s and thecivil rights movement was at the center of thewave in the 1960s Veterans organizations madegreat leaps forward during the 1930s and afterWorld War II persisting throughout the centu-ry but declining during the last half

The families next in coverage include SMOsfrom the feminist nativist and environmentalmovements (see Figure 2) The coverage of

feminist movement SMOs which in Figure 2also includes abortionreproductive rightsSMOs shows the expected two waves with thesecond wave beginning largely in the 1970sThe waves are fairly gentle however and thereis a ldquomiddlerdquo wave of coverage in the 1930s Thecoverage of environmental SMOs fits the pat-tern of a new social movement based on qual-ity-of-life concerns taking off in the 1970s and1980s peaking in the 1990s and sustaininghigh coverage By contrast nativist organiza-tions led mainly by two incarnations of the


Table 2 Times Coverage of SMOs by Movement Families

Without of Rank Family Title Percent Unions SMOs Most Highly Covered SMO

01 Labor 363 189 141 American Federation of Labor02 Civil Rights African American 98 124 62 National Association for the Advancement

mdashof Colored People03 Veterans 76 97 17 American Legion04 FeministWomenrsquos Rights 55 70 124 League of Women Voters05 NativistSupremacist 42 54 63 Ku Klux Klan06 EnvironmentConservation 40 51 132 Sierra Club

mdashEcology07 Civil Rights Jewish 37 47 7 American Jewish Congress08 Civil Liberties 31 40 6 American Civil Liberties Union09 Anti-war 28 36 79 American Friends Service Committee10 Conservative Other 26 33 98 John Birch Society11 Progressive Other 25 32 92 National Council of Jewish Women12 Anti-alcohol 24 30 21 Anti-Saloon League13 Farmers 21 26 18 American Farm Bureau Federation14 Communist 17 21 20 Communist Party USA15 Animal ProtectionRights 14 18 26 American Society for the Prevention of

mdashCruelty to Animals16 AbortionReproductive Rights 13 16 27 Planned Parenthood17 Civic 11 15 16 National Civic Federation18 Consumer 11 14 8 National Consumersrsquo League19 Old AgeSenior Rights 9 12 26 American Association of Retired People20 Christian Right 9 12 36 Moral Majority21 Civil Rights Other 9 11 34 Nation of Islam22 Childrenrsquos RightsProtection 9 11 13 Child Welfare League23 Liberal General 7 9 5 Americans for Democratic Action24 Lesbian Gay Bisexual and 5 6 47 Gay Menrsquos Health Crisis

mdashTransgender25 Anti-smoking 4 5 13 American Public Health Association26 Anti-abortion 4 5 33 National Right to Life Committee27 Gun Ownersrsquo Rights 3 4 11 National Rifle Association28 Civil Rights Native American 2 3 3 American Indian Movement29 Welfare Rights 2 2 12 National Welfare Rights Organization30 Civil Rights Hispanic 2 2 12 League of United Latin American Citizens31 Disability Rights 1 1 16 National Association for Retarded Children32 AIDS 1 1 5 AIDS Action33 Prison ReformPrisonersrsquo Rights 1 1 10 National Committee on Prisons34 Gun Control 1 1 14 Brady Campaign to Prevent Gun Violence


Wed 28 Oct 2009 035926


Figure 1 Times Coverage of Labor Movement African American Civil Rights Movement andVeterans SMOs 1900 to 1999

Figure 2 Times Coverage of the NativistSupremacist Feminist and Abortion Rights andEnvironmentalConservationEcology SMOs 1900 to 1999


Wed 28 Oct 2009 035926

KKK had a peak in coverage in the 1920s witha secondary peak in the 1960s

Across the twentieth century national news-paper coverage of SMOs focused on the laborand civil rights movements and scholarshiphas followed (eg Andrews 2004 Fantasia andStepan-Norris 2004 McAdam 1982 Morris1984) The labor movement has dominated cov-erage it remains the most covered movementfamily despite the precipitous decline in unionmembership in the last half of the twentiethcentury Similarly the African American civilrights feminist and environmental families ofSMOs rank expectedly high in coverage In arecent handbook (Snow et al 2004) a sectionon ldquomajorrdquo social movements included reviewsof the labor environmental and feminist move-ments and ethnic mobilization encompassingAfrican American civil rights and anti-warmovements but veterans and nativist move-ments were not covered Generally speakingSMOs that peaked in media attention beforethe 1960s and movements with a conservativeslant have not received scholarly attention com-mensurate with their media attention Whilethe top movement families also show waves ofcoverage as would be expected the coverageappears somewhat later than expected and is sus-tained longer than the imagery of cycles sug-gests

SIZE DISRUPTIVE ACTIVITY ANDCOVERAGE PRELIMINARY RESULTS

The descriptive results lead to the followingquestion Why are some SMOs and SMO fam-ilies better covered than others As noted ear-lier two approaches to the question are relatedto the scale of the movement and its activity Oneview is that newspapers disproportionately coverevents that are disruptive or violent (McCarthyet al 1996 Oliver and Myers 1999 see reviewin Earl et al 2004) and presumably SMOs con-nected to such events This view is connectedto the classic argument that disruption leads toinfluence for social movements (Piven andCloward 1977) One might also expect news-papers simply to report on SMOs according totheir size To some extent this is what reportersclaim to be doing (Gans 1979) and is consistentwith the resource mobilization view of theimpact of social movements (Zald andMcCarthy 2002) Movements are expected to

have influence in relation to available resourcesincluding the members and organizations in themovement family or industry These two aspectsof the scale of movements their size and dra-matic activity are frequently used to summarizeor operationalize the presence of movements andSMOs in quantitative research on movementsTo provide a preliminary assessment of thesemodels we compare newspaper coverage withmeasures employed in high-profile research onsome of the more prominent SMOs and SMOfamilies4

To address the degree to which coveragereflects the main aspects of SMO size we startwith two prominent SMOs The Townsend Planwas one of the most publicized SMOs of the1930s it demanded generous and universal old-age pensions and organized 2 million olderAmericans into Townsend clubs (Amenta et al2005) It quickly reached membership levelsthat few voluntary associations achieve (Skocpol2003) but it lost most of its following by the1950s The correlation between its membership(data from Amenta et al 2005) and coveragefrom 1934 to 1953 is 62 The NAACP a keyorganization in the most prominent movementof the second half of the twentieth century isby contrast an evergreen in coverage In exam-ining data from 1947 through 1981 (courtesy ofJ Craig Jenkins) we find the relationshipbetween its membership and Times coverage isfairly strong too with a correlation of 69Membership and coverage both peak in themid-1960s

We next address the connection between cov-erage and size for two of the most prominentSMO families beginning with organizationaldensity in the womenrsquos rightsabortion rightsmovements from 1955 through 1986 (with data


4 These models are similar to debates in the liter-ature on newspaper coverage of protest events whichseek to uncover selection and description biases in thecoverage (see review in Earl et al 2004) Factorsmaking events seem more newsworthy include prox-imity to news organizations size intensity presenceof violence counter-demonstrators police and spon-sorship by organizations Unlike some studies ofselection bias of protest events (McCarthy et al1996 Oliver and Myers 1999) our preliminary inves-tigations of SMO coverage cannot juxtapose all rel-evant aspects of size or activity of SMOs with theircoverage as data on these aspects do not exist


Wed 28 Oct 2009 035926

courtesy of Debra C Minkoff) A plot of SMOcoverage in a three-year moving averageagainst the organizational density of total organ-izations and the subset of ldquoprotest and advoca-cyrdquo organizations in the womenrsquos rightsmovement shows that they are very stronglyand similarly correlated (97) (see Figure 3)Coverage and organizational density both risedramatically in the mid-1960s and peak around1980 Despite the large correlation betweencoverage and organizational density howeveronly a few SMOs received the bulk of the cov-erage As for the most prominent family a com-parison of the Times coverage of the labormovement from 1930 to 1999 with unionizationshows a correlation of 59 after 1954 howev-er when unionization declines the correlationincreases to 80 (see Figure 4)

Next we turn to bivariate assessments ofwhether coverage is closely connected to dis-ruptive activity We begin with labor strikesthe standard disruptive activity of the labormovement (see Figure 4) The pattern for cov-erage and strikes works in the opposite directionfrom unionization Although the correlationbetween the work stoppage measures and cov-

erage is 58 overall between 1930 and 1947during the rise of the labor movement the cor-relation is 815 In short correlations are highfor strike activity in the early years of the labormovement and high for unionization in lateryears Coverage may generally result from dis-ruptive action in the early years of a largelysuccessful movement and from aspects of itssize in later years

Next we assess the connection between cov-erage and protest events in the African Americancivil rights movement the second most cov-ered movement family Jenkins Jacobs andAgnone (2003286) extending McAdamrsquos(1982) data for 1950 through 1997 defineprotest events as ldquononviolent protest by AfricanAmericans including public demonstrationsand marches sit-ins rallies freedom rides boy-


Figure 3 The Density of Feminist and Abortion Rights Organizations by Overall andProtestAdvocacy Organizations and Times Coverage 1955 to 1986

5 In the Bureau of Labor Statisticsrsquo data for theyears 1947 to 1999 ldquowork stoppagerdquo includes onlythose involving at least 1000 workers whereas ear-lier data include work stoppages of any number ofworkers In the 15 years in which the two measuresoverlap they have a correlation of 96


Wed 28 Oct 2009 035926

cotts and other protest actionsrdquo We comparethis measure with coverage of the so-called BigFour civil rights organizations the NAACP theStudent Nonviolent Coordinating Committee(SNCC) the Congress of Racial Equality(CORE) and the Southern Christian LeadershipConference (SCLC) As Figure 5 shows thetwo have the same general pattern with smallincreases in the late 1950s followed by largerincreases in the 1960s and a relatively con-stant and low level of activity starting in the1980s they are correlated at 66 Although bothcoverage and protest events level out after theearly 1970s coverage has remained at a fairlyhigh level despite far fewer protests

All in all these preliminary bivariate resultsshow that coverage tracks to some degree SMOand SMO family size as well as disruption anddramatic activity The medium high correla-tions between coverage and individual mem-bership for two prominent SMOs in conjunctionwith higher correlations with union density anda very high correlation with feminist SMOssuggest that coverage is connected most close-ly to the size of entire influential movementfamilies Approximately 43 percent of thenational SMOs we located typically small

organizations gain little or no coverage Thissuggests that size matters coverage generallyconcentrates on the better-known SMOs inmovement families These findings are consis-tent with the resource mobilization view ofmovementsrsquo impact Coverage is also relatedto protest and similar activity especially in theearly days of a movement organization or fam-ily For SMOs and SMO families that do notgain organizational footholds after early yearsof disruptive or dramatic activities the earlydays are all they have In short the preliminaryresults indicate some support for both disrup-tion and resource mobilization explanations ofmovement outcomes These two views howev-er are not inconsistent with each other and wefurther test them below

WHY ARE SOME SMO FAMILIESBETTER COVERED THAN OTHERS

We now turn to systematic comparative analy-ses of coverage across SMO families To addresswhy some movement families received exten-sive coverage in their careers we employ fuzzy-set qualitative comparative analysis (fsQCA)Relying on set logic fsQCA is typically used to


Figure 4 Union Membership Work Stoppages and Times Labor Movement SMO Coverage 1930to 1999

Work Stoppages (All)


Wed 28 Oct 2009 035926

examine unusual occurrences (see Ragin 19872000) Instead of focusing on how much a givenmeasure adds to explained variance fsQCAaddresses the possibility that causes are con-juncturalmdashthat is two or more conditions mustoccur simultaneously to produce a result It alsoaddresses the possibility of multiple causa-tionmdashthat more than one conjunctural causalpath will lead to a result High coverage isindeed an unusual occurrence and we expecthigh coverage to result from multiple causes Weseek to develop an explanation inductively byusing ideas and measures from the main macrotheories of the development and impact of socialmovements Set-theoretical thinking and analy-ses are especially appropriate here because thesetheories are often treated as complementaryrather than competing (McAdam 1996)

To identify potential determinants of cover-age at the SMO family level we go beyond thedisruption and resource mobilization modelsand address two arguments about the influenceof political contexts on movements and the out-comes they seek to affect The first concerns the

impact of political partisanship the central polit-ical context most often held to influence move-ments and their consequences (Meyer andMinkoff 2004) We also address a second andless frequently analyzed political contextwhether an SMO family benefits from anenforced national policy benefiting its con-stituents (Amenta and Young 1999 Berry 1999see also Baumgartner and Jones 1993Halfmann et al 2005)

OUTCOME AND CAUSAL MEASURES

We focus on daily coverage an SMO familyreceiving one mention or more per day in theNew York Times Among the movement familiesreaching daily coverage for at least one year dur-ing the past century are the anti-alcohol anti-war environmental feminist old-age nativistand veterans movements Most incidences ofyearly daily coverage involve the two most pub-licized movement familiesmdashthe labor move-ment and the African American civil rightsmovement which received at least daily cover-


Figure 5 Times Coverage of the African American Civil Rights Big Four SMOs and ProtestEvents 1950 to 1997

Note The African American civil rights Big Four SMOs are the NAACP the Student Nonviolent CoordinatingCommittee (SNCC) the Congress of Racial Equality (CORE) and the Southern Christian Leadership Conference(SCLC)


Wed 28 Oct 2009 035926

age from 1919 through 1999 and 1960 through1981 respectively These two movements werealso the only ones ever to achieve coveragetwice a day a level we analyze separately Threeother SMO families achieved stretches of dailycoverage lasting five years or longer the vet-erans movement (1921 through 1941 1945through 1952) the anti-alcohol movement (1926through 1931) and the environmental move-ment (1982 through 1993) These strings ofcoverage make up about 90 percent of the cases(movement-family-years) of daily coverageThese families also come close to achievingdaily coverage before and after their strings ofdaily coverage (To smooth out spikes in cov-erage we measure it by way of a three-yearmoving average) Several SMO families fallsomewhat short of ever receiving daily cover-age including the farmers in the 1930sCommunist SMOs in the 1930s Jewish civilrights in the 1950s 1960s and 1970s civil lib-erties in the 1970s and 1980s and the Christianright in the 1990s Many of these SMO fami-lies did however achieve coverage every otherday which is still very high and which we ana-lyze separately

Our unit of analysis is the SMO family-yearEach of the 31 movement families receives ascore for each year of the twentieth century forcoverage measured by number of articles eachcausal measure is tracked similarly Needless tosay not all 31 SMO families were in existencethroughout the century We consider a familyrsquosfirst appearance to have occurred once twoSMOs in it were founded yielding 2153 fam-ily-year observations Coverage (C) scores onefor each year in which an SMO family receiveddaily or more frequent coverage (ie scoresone for 365 or greater mentions) In crisp setfsQCA models each measure is categoricalwith a score of one or zero Approximately 76percent of the 2153 SMO family years experi-enced daily or greater coverage In some of theanalyses however we examine twice-per-daycoverage (34 percent of the cases achieve 730or greater mentions) and every-other-day cov-erage (161 percent achieve 1825 or greatermentions)

As for the causal conditions we developmeasures from the four main perspectives out-lined above We expect that a combination ofthree or more of these four conditions may needto occur simultaneously to explain why and

when some movement industries receive exten-sive coverage For disruption (D) a family yearscores one if any organization in the SMO fam-ily was engaged in either illegal collective actionor disruptive action such as strikes boycottsoccupations and unruly mass protests that drewthe reaction of authorities or collective actionin which violence was involved whether by themovement authorities or opponents of themovement We generated the scores from schol-arly monographs about the families and Websites of current organizations For the resourcemobilization model we score one if 30 or moreorganizations were ldquoactiverdquo in a given year Forthis measure of organizations (O) organiza-tions are considered active after their date ofbirth which we established from scholarlymonographs and Web sites of current organi-zations The actual yearly counts of all organi-zations in all SMO families are of courseunknown but the measure is not derived fromcoverage figures and includes many organiza-tions never covered

The first political contextual measure par-tisanship (P) scores one for non-conservativeSMO families each year in which a Democratwas president with a Democratic majority inCongress for conservative movement familiesthis measure scores one for Republican presi-dents with Republican majorities (Poole andRosenthal 2008) A second political contextu-al measure enacted and enforced policy (E)scores one for years after the enactment of amajor policy in favor of the movement familyrsquosissue or main constituency provided that anational bureau or department was in place toenforce or administer the law (Aberbach andPeterson 2005 Baumgartner and Jones 2008)In all 159 percent of the cases are coded pos-itive on disruption 109 on organizations 292on partisanship and 321 percent on enforcedpolicy

RESULTS AND DISCUSSION

Set analyses such as fsQCA can identify limit-ed diversity among causal conditions in datasets Ideally there would be nearly equal dis-tribution across causally relevant measures butthis condition rarely holds in the non-experi-mental studies typical in historical social sci-ence although researchers often act as thoughit were otherwise (Ragin 2000 2008) Because



Wed 28 Oct 2009 035926

there are four causal conditions the truth table(see Table 3) has 16 (or 24) potential combina-tions None are completely empty but somehave many more cases than others The largestnumber of cases 881 falls into the category inwhich all causal conditions are absent Similarly851 cases fall into the four combinations forwhich all but one of the causal conditions iscoded as absent As we will see these five com-binations (dope Dope dOpe doPe and dopE)rarely coincided with extensive newspaper cov-erage at any of the three levels From the otherdirection where the data are sparse three of 16combinations (DOPe DoPE dOPe) made upfewer than 22 cases or less than 1 percent of thecases Unless otherwise indicated we elimi-nated from the analyses these very low fre-quency combinations treating them as negativecases6

As a preliminary to the fsQCA analyses weran a random-effects negative binomial regres-sion model of coverage (using raw coveragefigures) with the 2153 issue-years serving asthe units of analysis on the four major meas-ures plus dummy measures for each year(Greene 2007 Long and Freese 2005Wooldridge 2002) The results (not shownavailable on request) indicate that each of theindependent measures has a positive effect oncoverage Coefficients for each are significantat the 01 level These positive results howev-er may largely be a function of the fact that somany of the cases reside in the no causeno out-come cells Also we expect the factors to worklargely in combination to produce high cover-age

To address the combinations of character-istics that led to daily coverage for movementfamilies we examine the rows of the truthtable in which all or a significant majority ofthe cases (at the 05 level) are positive weeliminate combinations with less than 1 percentof the cases We employ fsQCA 20 (RaginDrass and Davey 2006) augmented by the


Table 3 Four-Measure FsQCA Crisp Outcomes and Top SMO Families by Combination

Outcome Success Total Most Prominent SMO Families (successestotal cases)

DOPE 35 39 Labor (3030) CR-African American (44) Environment (15)DOpE 54 63 Labor (3535) Environment (1417) Feminist (06) Anti-abortion (06)

mdashCR-African American (55)DoPE 5 7 Veterans (44) AIDS (02) CR-African American (11)DOPe 3 6 Anti-war (14) Labor (22)dOPE 4 16 Feminist (09) CR-African American (46) Environment (01)DOpe 15 42 NativistSupremacist (021) Labor (1414) Anti-war (17)dOpE 8 39 CR-African American (820) Feminist (019)DoPe 6 41 NativistSupremacist (021) CR-Jewish (07) Labor (04) CR-African

mdashAmerican (33) Animal ProtectionRights (02)doPE 20 157 Farmers (036) Old Age (030) Veterans (1128) Feminist (323) Anti-alcohol

mdash(612)DopE 1 31 Anti-abortion (011) AIDS (08) CR-American Indian (06) CR-Hispanic (05)

mdashVeterans (11)dOPe 0 11 Anti-war (010) LGBT Rights (01)Dope 3 118 NativistSupremacist (054) CR-Jewish (025) Labor (015) Animal

mdashProtectionRights (08) CR-African American (36)dopE 4 349 Childrenrsquos Rights (088) Farmers (064) Veterans (437) Old Age (035)

mdashFeminist (023) Consumer (022)dOpe 0 23 Anti-war (018) LGBT Rights (05)doPe 0 361 Animal ProtectionRights (034) Communist (032) Environment (030)

mdashCR-Jewish (029) Disability Rights (029) Prison Reform (029)dope 9 881 Civic (0100) Anti-smoking (092) Anti-alcohol (074) Christian Right (065)

mdashAnimal ProtectionRights (056) Communist (048)

Note CR = civil rights ldquoDrdquo is a measure of disruptive capacity ldquoOrdquo is a measure of organizations ldquoPrdquo is ameasure of partisan political context and ldquoErdquo is a measure of enforced policy See text for operationalizations

6 In none of the four small-N combinations hereare the high-coverage cases greater in number thanthe nonndashhigh-coverage cases This is not true how-ever for some of the analyses below


Wed 28 Oct 2009 035926

Stata 101 fuzzy command (Longest and Vaisey2008) which provides probabilistic statisticaltests We also adjust standard errors for intra-movement correlations We locate two combi-nations for which the positive cases aresignificantly greater than the negative cases atthe 05 level DOPE for which all the inde-pendent measures are present and 36 of 39cases are positive and DOpE for which onlythe political partisanship measure is absentand 54 of 63 cases score positive (In fsQCAterminology the presence of a causal conditionis indicated by upper case and its absence bylower case a plus sign [+] indicates the oper-ator ldquoorrdquo or set union and an asterisk [] indi-cates the operator ldquoandrdquo or set intersection)Through the use of Boolean algebra thesecombinations reduce to the following result C= DOE

This result means that daily coverage isexplained by the joint presence of disruptionorganizations and an enforced policy Partisanalignment is not part of the solution and thereis no additional solution The solution is con-junctural but not multiple This solution ldquocov-ersrdquo 533 percent of the dependent measurecases with a ldquoconsistencyrdquo of 873 percent InBoolean or set logic terms ldquoconsistencyrdquo meansthe degree to which cases with a given combi-nation of causal conditions constitute a subsetof the cases with the outcome ldquoCoveragerdquo indi-cates the degree of overlap between the caseswith the causal combination and the cases withthe outcome For our result above one canimagine a Venn diagram in which a set formedby the intersection or overlap of the sets of thethree causal measures (D O and E) in turnoverlaps slightly more than one half with theoutcome set (C) at the same time less than 13percent of the cases with the causal combina-tion fall outside the set of cases with the out-come To deploy a more graphic if gruesomeUS example very similar to the results aboveusing a gun in a suicide attempt is ldquoconsistentrdquowith achieving a ldquosuccessfulrdquo suicide at a rateof about 89 percent gun-initiated suicides alsoaccounted for or ldquocoveredrdquo about 54 percent ofsuicides in 2005 (Anderson 2008)

We also reran the analyses using fuzzy ratherthan crisp sets for the outcome measure C andfor causal condition O the number of activeorganizations There are many analytical advan-tages to fuzzy sets (Ragin 2000) Unlike crisp

sets which employ categorical measures fuzzysets indicate the degree to which a case hasmembership in a set using the same set logicfuzzy sets can exploit greater variance in meas-ures to designate degrees of membership insets For instance in the crisp set analyses anySMO family that had 364 days of coverage ina given year would be considered completelyoutside the set of daily coverage With fuzzysets however this family year would be con-sidered almost entirely inside the set of dailycoverage This matters because SMO familiesoften scored just below achieving daily cover-age before and after their strings of daily cov-erage and as noted several SMO familiessometimes scored close to daily coverage Todevise fuzzy sets researchers must choose aceiling above which a measure is considered tobe ldquofully inrdquo the setmdashusually the same as thecutoff point for crisp setsmdashand a floor belowwhich a measure is considered to be ldquofully outrdquoof the set Here we use the ldquodirect methodrdquo forderiving partial membership scores (Ragin2008)7 In our case we code coverage of oncea day or more frequently as fully in the set ofdaily coverage and coverage of once a week(52) or fewer mentions per year as fully out ofthe set As for the organizations measure wecount a family with 30 or more organizations inexistence as fully in the set of high organiza-tions while five organizations or fewer indicatefully out status The other three independentmeasures remain categorical

The fuzzy set results confirm the crisp setresults for daily coverage Again using the cri-terion for selection as positive scores being sig-nificantly greater than negative scores at the05 level and eliminating any combinationswith less than 1 percent of the cases we locatethe same two truth table combinations as for thecrisp sets The reduced result is the same C =DOE This solution covers less of the fuzzyset outcome (22 percent) but is more consistentwith it than the crisp set result (at 913 per-cent) These differences are not surprising as theset of daily coverage expands by making itfuzzy The results support the disruption andresource mobilization arguments as well as the


7 This means that partial membership scores in theset are computed based on deviations from thesethresholds on a log odds scale (Ragin 2008)


Wed 28 Oct 2009 035926

political contextual theory invoking enforcedpolicies

We also checked whether the results hold forthe post-World War II period when the New YorkTimes sought to be a truly national newspaperUsing the same standards as before the solutionis the same C = DOE In each case the solu-tion has similar levels of consistency althoughwith increased rates of coverage For the crispset analyses the level of consistency is approx-imately the same at 859 percent whereas thecoverage increases to 731 percent For the fuzzyset analyses the solution consistency is approx-imately the same at 904 percent and the cov-erage increases similarly to 288 percent8 Inshort the results hold for the postwar period andprovide a better fit

With fsQCA it is also possible to explainnegative cases that is movement family yearswhen coverage was not extensive Unlike regres-sion methods set theoretical analyses do notassume that the causes of negative and positiveoutcomes are parallel (Ragin 2008) Using thesame standards of significance as above wefind that 10 truth table combinations have sig-nificantly negative scores Solving for the neg-ative combinations for the crisp sets yieldsc = do + dp + de + op + oe Thesolution covers 305 percent of the casesat a solution consistency of 975 per-cent This result can be better under-stood as c = d(o + p + e) + o(d + p + e) Thismeans that the absence of disruptive capacitiesand the absence of any one of the other threeconditions lead to non-daily coverage so toodoes the absence of a high number of organi-zations and the absence of any one of the otherthree conditions For the fuzzy set analysesthis same solution covers 592 percent of thecases at 856 percent consistency9

We also examined two other outcomemeasures twice-a-day coverage and every-other-day coverage Only the labor AfricanAmerican civil rights and veterans move-ments ever received the extremely high rateof coverage The best solution for both crispand fuzzy set analyses includes all four ofthe causal conditions C = DOPE Thecrisp results cover 432 percent of the set ata rate of 821 consistency whereas the fuzzyset combination covers 18 percent at 914percent consistency10 For the fuzzy set analy-ses however the combination DOpE is pos-itive and falls just short of significance If thisis treated as a positive combination the solu-tion is the familiar C = DOE which cov-ers 423 percent of the outcome cases at a 798percent level of consistency In their heydaysthe labor and civil rights movements includedall of the four determinants of high coverageThese families were characterized by disruptivecollective action and large numbers of organi-zations and each benefited from the Democraticregimes of the 1930s and 1960s Each alsogained key concessions during these periodsfavorable policies were enacted with consider-able bureaucratic enforcement

Finally we examined every-other-day cover-age These analyses were prompted by the factthat the bulk of SMO families can reasonablyaspire to somewhat less press than daily cover-age For crisp sets an SMO family requires1825 days or greater of coverage and the firstresult is the same as the daily result C = DOEThis has somewhat lower coverage (285 per-cent) given that more SMO family years qual-ify although its consistency increases to971 percent11 For the more accurate fuzzy


8 To check the robustness of the main result weengaged in a series of other analyses Using raw cov-erage does not change the overall resultmdashthe consistencyis the same and the coverage is slightly highermdashand nei-ther does eliminating any single movement family (notshown results available on request) The only movementfamily whose removal influences the results is laborlargely because of a loss of cases in the cells where manyof the causal conditions are present The results how-ever remain consistent (not shown results available onrequest)

9 For the fuzzy sets however one of the combinationsis not significant at the 10 level When this one is elim-

inated the solution becomes c = dp + de + op + oeThis so lu t ion which can be rewri t ten as c = d(p + e) + o(p + e) covers 50 percent of the casesat 870 percent consistency In this instance one set ofpaths goes through the absence of disruption and theabsence of either of the political contextual measuresthe second set involves the absence of organizations andthe absence of either of the political contextual meas-ures

10 For fuzzy sets 730 days of coverage counts as fullyin the set and five days per week coverage (260 days)counts as fully out with ldquodirect transformationsrdquo for par-tial membership

11 Two small-N combinations however are signifi-cant at the 10 level When we include these two com-

In order toget the equa-tions to print

all on oneline as you

requested inyour priormark-upexcessive

letterspacingwas required

before andafter in some

situationsFixing the

spacing andhaving the

equations onone line are

mutuallyexclusiveunless wemake the

equationsdisplay

equations

The extra

space thatwill require

will cause re-flow that will

result in anextra page

thus creatingthe need tore-paginate

the remainderof the issue


Wed 28 Oct 2009 035926

set analyses the results are similar12 Thetypical standards of signif icance and num-bers of cases produce the following solu-tion C = DOE + DOp As with the crispset results the greatest amount of coverage isprovided by the DOE term (161 percent cov-erage 946 percent consistency) with the solu-tion as a whole providing 280 percent coverageat an 867 percent level of consistency In thislast solution disruption and a high number oforganizations appear in both terms13

All in all the results form a consistent pat-tern as Table 4 indicates For daily coverage the

main findings indicate that disruption organi-zations and an enforced policy are togethersufficient These findings are strengthened whenwe confine the period of coverage to the post-war period when the Times became more devot-ed to national issues To explain the highestlevel of coverage the solution includes thesimultaneous occurrence of all the causal con-ditionsmdashdisruption organizations partisanregimes and enforced policies When we reducethe standard to every-other-day coverage thissolution of disruption organizations andenforced policy remains the dominant one Eachof the perspectives receives support from thefsQCA analyses and no one factor is a magicbullet that produces coverage The strongestsupport goes to the resource mobilization the-ory Moreover disruption appears in almost allsolutions as does the enforced policy measurebased on political contextual arguments cen-tering on the adoption of policies The partisanalignment factor however figures only in thesolution for the highest amount of coverage

CONCLUSIONS

Social movement organizations are crucial topolitical life and media coverage of SMOs iskey to both substantiating their claims to rep-resent groups and developing important cul-tural outcomes This article documents thenational newspaper coverage received by nation-


binations the result is C = O(DE + DP + PE) Thissolution includes the standard term DOE plus theterms DOP and OPE and covers 334 percentof the cases with a 935 level of consistency The pathsDOP and OPE thus add 49 percentage pointsto the coverage This result suggests that a high num-ber of organizations matter the most and appear ineach of the three solutions but any two of the otherconditions are also required

12 For fuzzy sets every-other-day coverage (1825days per year) counts as fully in and one day per week(26 days) counts as fully out of the set with standardtransformations for partial membership

13 A small-N combination is significantly positiveat the 05 level an additional larger-N combinationis significant at the 10 level and yet another small-N combination has a much higher ldquoyesrdquo consisten-cy than ldquonordquo consistency When entering all but thelast case as true and treating that one as ldquodonrsquot carerdquo(Ragin 2008) the solution is C = DO + OPE Thissolution covers 377 percent at an 848 percent levelof consistency As before a large number of organ-izations are a part of each element of the solutionwith one path including disruption and the other the

Table 4 Four-Measure FsQCA Solutions for Extensive Coverage by Amount of Coverage and Typeof Analysis

Amount of Type of True Reduced Solution Solution Coverage Analysis Combinations Solution Coverage Consistency

Daily Crisp DOPE DOpE DOE 533 873Daily Fuzzy DOPE DOpE DOE 220 913Daily (Postwar) Crisp DOPE DOpE DOE 731 859Daily (Postwar) Fuzzy DOPE DOpE DOE 288 904Twice Per Day Crisp DOPE DOPE 432 821Twice Per Day Fuzzy DOPE DOPE 180 914Every Other Day Crisp DOPE DOpE DOE 285 971Every Other Day Fuzzy DOPE dOPE DOE + 354 849

Dope DOpE DOp

Note See text for operationalizations significance levels and additional results

combined influence of the political contextual con-ditions


Wed 28 Oct 2009 035926

al US SMOs and families in the twentieth cen-tury as a prelude to explaining why some fam-ilies were extensively covered Our analysesprovide the first test of the main social move-ment theories including those regarding dis-ruption resource mobilization and politicalcontexts across all movements on a measure ofthe cultural influence of movements

Our fsQCA analyses of daily or greater cov-erage by social movement families or indus-tries show some support for the mainmacro-social theories of social movements andtheir consequences The results indicate thatextremely high coverage at the level of twicea day is best explained by each of the fourdeterminantsmdashdisruptive activity a large num-ber of organizations a favorable politicalregime and an enforced policy in favor of theSMO familyrsquos constituencymdashoccurring at thesame time In their heydays the labor andAfrican American civil rights movements hadthis sort of saturation coverage To producedaily coverage we find that short-term partisancontexts are not important the main solutionincludes only disruption large numbers oforganizations and an enforced policy This com-bination is also a main part of the solution forevery-other-day coverage Most solutionsinclude disruption and a large number of organ-izations in existence In combination with thebivariate analyses these set-theoretic resultsprovide strong support for the resource mobi-lization and disruption arguments which seemto work in tandem to influence coverage

The results also suggest though that schol-ars need to rethink their ideas regarding whatconstitutes a favorable political context formovements Enforced policies seem to mattermore for coverage at least than do favorablepartisan circumstances It is possible howeverthat these causes work sequentially and thathighly partisan contexts are critical for the devel-opment of new policies in favor of movementsrsquoconstituencies The Democratsrsquo partisan domi-nance in the mid-1930s and 1960s was closelyconnected to new policy developments For thelabor movement these policies centered on theWagner Act of 1935 and the creation of theNational Labor Relations Board the AfricanAmerican civil rights movement saw the CivilRights Act of 1964 and the Voting Rights Actof 1965 Similarly dramatic action may havebeen important in spurring these sorts of poli-

cies which provided SMOs with both politicaland cultural leverage Policy-related controver-sies may help keep SMO families in the newsand in public discourse long after their disrup-tive peaks

Like the initial analyses of the political con-sequences of social movements however ouranalyses and results which examine the broad-est macro perspectives about the causes andconsequences of social movements are onlythe first steps in theorizing and analyzing theprocess of gaining coverage Analyses of polit-ical outcomes have moved beyond movement-centered models and theorized more extensivelyinteractions between movements and politicalstructures and processes (see Amenta 2006Andrews 2004) Similarly more complete the-orizing of interactions between movements andmedia structures and processes will likely pro-vide more compelling theoretical claims andmore accurate analyses of SMO coverageMoreover coverage is a limited measure ofinfluence for SMOs and SMO families Rawcoverage does not identify whether an SMOachieved ldquostandingrdquo nor whether articles includ-ed frames favorable to a movement or if thetone or valence of coverage was favorable Alsowinning discursive battles in newspapers doesnot necessarily translate into favorable policyoutcomes for social movements (Ferree et al2002) Examining coverage in a more refinedway and connecting it with thinking and analy-ses of policy outcomes is needed to establishthe nature of these links

Our descriptive and bivariate findings alsohave implications for further inquiry Coverageof movements corresponds in part with previ-ous scholarly attention to movements Labormovement organizations and similarly well-studied African American civil rights SMOsare best covered Yet veterans nativist and civilliberties SMOs received coverage that far out-strips corresponding scholarship and general-ly speaking SMOs from before the 1960s andnon-left SMOs (see McVeigh 2009) are not aswell researched as they are covered Possibly dif-ferent theoretical claims will apply to them Inbivariate analyses we find that newspaper cov-erage closely reflects movement size at theSMO family level for the prominent labor andfeminist movements and larger SMOs receivefar more coverage The results also show thatcoverage tracked strikes and protest events dur-



Wed 28 Oct 2009 035926

ing the rise of the labor and African Americancivil rights movements Coverage thus seems totrack conspicuous collective action in the earlyyears of an SMO or SMO family followed bycoverage according to size for older organiza-tions at least for some highly influential SMOfamilies This pattern corresponds to ideas aboutthe institutionalization of movements (Meyerand Tarrow 1998) but it may apply only toSMO families that achieve permanent leveragein politics

These results suggest a few additional newdirections in research It would be revealing tocompare the newspaper coverage of well-stud-ied SMOs with a wide range of their actionsanalogous to work on protest and its coverageto ascertain which activities and characteristicsof SMOs tend to lead to coverage and which donot In regressing measures of size and activi-ty on coverage with various control measuresmoreover it may be possible to devise ways toadjust coverage figures so that they more close-ly tap these less-easily measured aspects ofSMOs and SMO families These adjusted meas-ures could be valuable in addressing many ques-tions about social movements and this line ofresearch may hasten the day when analysesacross movements and over time will no longerseem exceptional

Edwin Amenta is Professor of Sociology and Historyat the University of California-Irvine He is the authormost recently of When Movements Matter TheTownsend Plan and the Rise of Social Security(Princeton 2008 paperback) and Professor BaseballSearching for Redemption and the Perfect Lineup onthe Softball Diamonds of Central Park (Chicago2007)

Neal Caren is an Assistant Professor of Sociology atthe University of North Carolina-Chapel Hill Hisresearch interests center on the quantitative andqualitative comparative analysis of protest and socialmovements the intersection of place and politicalaction and the environmental justice movement andpollution disparities

Sheera Joy Olasky is a PhD candidate in sociologyat New York University Her research interests arecentered on political sociology and social move-ments She is coauthor (with Edwin Amenta and NealCaren) of ldquoAge for Leisure Political Mediation andthe Impact of the Pension Movement on US Old-AgePolicyrdquo ASR 2005

James E Stobaugh is a PhD candidate in sociolo-gy at the University of California-Irvine His inter-ests are in social movements and political sociology

and his research addresses the intersection of reli-gion law and society His masterrsquos thesis examinesthe framing of the creationist and intelligent designmovements His other research examines the mediacoverage of US social movements

REFERENCES

Aberbach Joel D and Mark Peterson 2005Institutions of American Democracy The ExecutiveBranch New York Oxford University Press

Amenta Edwin 2006 When Movements MatterThe Townsend Plan and the Rise of Social SecurityPrinceton NJ Princeton University Press

Amenta Edwin Neal Caren and Sheera Joy Olasky2005 ldquoAge for Leisure Political Mediation andthe Impact of the Pension Movement on US Old-Age Policyrdquo American Sociological Review70516ndash38

Amenta Edwin and Michael P Young 1999ldquoDemocratic States and Social MovementsTheoretical Arguments and Hypothesesrdquo SocialProblems 46153ndash68

Anderson Scott 2008 ldquoThe Urge to End It All IsSuicide the Deadly Result of a Deep PsychologicalConditionmdashor a Fleeting Impulse Brought on byOpportunityrdquo New York Times Magazine July 6pp 38ndash43

Andrews Kenneth T 2004 Freedom Is a ConstantStruggle The Mississippi Civil Rights Movementand Its Legacy Chicago IL University of ChicagoPress

Andrews Kenneth T and Bob Edwards 2004ldquoAdvocacy Organizations in the US PoliticalProcessrdquo Annual Review of Sociology 30479ndash506

Baumgartner Frank R and Bryan D Jones 1993Agendas and Instability in American PoliticsChicago IL University of Chicago Press

mdashmdashmdash 2008 Center for American Politics andPublic Policyrsquos Policy Agendas Project(httpwwwpolicyagendasorg)

Berry Jeffrey M 1999 The New Liberalism TheRising Power of Citizen Groups Washington DCBrookings Institution

Brulle Robert Liesel Hall Turner Jason Carmichaeland J Craig Jenkins 2007 ldquoMeasuring SocialMovement Organization Populations AComprehensive Census of US EnvironmentalMovement Organizationsrdquo Mobilization AnInternational Journal 12255ndash70

Clemens Elisabeth S 1997 The Peoplersquos LobbyChicago IL University of Chicago Press

Corbett Julia B 1998 ldquoMedia Bureaucracy andthe Success of Social Protest Newspaper Coverageof Environmental Movement Groupsrdquo MassCommunications and Society 141ndash61

Cress Daniel and David A Snow 2000 ldquoTheOutcomes of Homeless Mobilization TheInfluence of Organization Disruption Political



Wed 28 Oct 2009 035926

Mediation and Framingrdquo American Journal ofSociology 1051063ndash1104

Earl Jennifer 2004 ldquoCultural Consequences ofSocial Movementsrdquo Pp 508ndash30 in The BlackwellCompanion to Social Movements edited by D ASnow S A Soule and H Kriesi Malden MABlackwell Publishing

Earl Jennifer Andrew Martin John D McCarthy andSarah A Soule 2004 ldquoThe Use of NewspaperData in the Study of Collective Actionrdquo AnnualReview of Sociology 3065ndash80

Fantasia Rick and Judith Stepan-Norris 2004 ldquoTheLabor Movement in Motionrdquo Pp 555ndash75 in TheBlackwell Companion to Social Movements edit-ed by D A Snow S A Soule and H KriesiMalden MA Blackwell Publishing

Ferree Myra Marx William Anthony GamsonJurgen Gerhards and Dieter Rucht 2002 ShapingAbortion Discourse Democracy and the PublicSphere in Germany and the United StatesCambridge UK Cambridge University Press

Fountain Kathi Carlisle 2006 ldquoPolitical AdvocacyGroups Alphabetic Listrdquo (httpwwwvancouverwsuedufackfountain)

Gamson William A 1990 The Strategy of SocialProtest 2nd ed Belmont CA WadsworthPublishing

mdashmdashmdash 1998 ldquoSocial Movements and CulturalChangerdquo Pp 57ndash77 in From Contention toDemocracy edited by M G Giugni D McAdamand C Tilly Landham MD Rowman andLittlefield

mdashmdashmdash 2004 ldquoBystanders Public Opinion andthe Mediardquo Pp 242ndash61 in The BlackwellCompanion to Social Movements edited by D ASnow S A Soule and H Kriesi Malden MABlackwell Publishing

Gamson William A David Croteau WilliamHoynes and Theodore Sasson 1992 ldquoMediaImages and the Social Construction of RealityrdquoAnnual Review of Sociology 18373ndash93

Gans Herbert 1979 Deciding Whatrsquos News A Studyof CBS Evening News NBC Nightly NewsNewsweek and Time New York Pantheon Books

Ganz Marshall 2000 ldquoResources andResourcefulness Strategic Capacity in theUnionization of California Agriculture1959ndash1966rdquo American Sociological Review1051003ndash1062

Gitlin Todd 1978 ldquoMedia Sociology The DominantParadigmrdquo Theory and Society 6205ndash53

mdashmdashmdash 1980 The Whole World is Watching MassMedia in the Making and Unmaking of the NewLeft Berkeley CA University of California Press

Granados Francisco J and David Knoke 2005ldquoOrganized Interest Groups and Policy NetworksrdquoPp 287ndash309 in Handbook of Political SociologyStates Civil Societies and Globalization editedby T Janoski R R Alford A M Hicks and M

A Schwartz New York Cambridge UniversityPress

Greene William H 2007 Econometric Analysis 6thed Upper Saddle River NJ Prentice Hall

Halfmann Drew Jesse Rude and Kim Ebert 2005ldquoThe Biomedical Legacy in Minority HealthPolicy-Making 1975ndash2002rdquo Research in theSociology of Health Care 23245ndash75

Jenkins J Craig and Craig Eckert 1986ldquoChallenging Black Insurgency Elite Patronageand Professional Social Movement Organizationsin the Development of the Black MovementrdquoAmerican Sociological Review 51812ndash29

Jenkins J Craig David Jacobs and Jon Agnone2003 ldquoPolitical Opportunities and African-American Protest 1948ndash1997rdquo American Journalof Sociology 109277ndash303

Kerbo Harold R and Richard A Shaffer 1992ldquoLower Class Insurgency and the Political ProcessThe Response of the US Unemployed1890ndash1940rdquo Social Problems 39139ndash54

Koopmans Ruud 2004 ldquoMovements and MediaSelection Processes and Evolutionary Dynamicsin the Public Sphererdquo Theory and Society33367ndash91

Lipsky Michael 1968 ldquoProtest as a PoliticalResourcerdquo The American Political Science Review621144ndash58

Long J Scott and Jeremy Freese 2005 RegressionModels for Categorical Dependent Variables UsingStata 2nd ed College Station TX Stata Press

Longest Kyle C and Stephen Vaisey 2008 ldquoFuzzyA Program for Performing QualitativeComparative Analyses (QCA) in Statardquo The StataJournal 879ndash104

McAdam Doug 1982 Political Process and theDevelopment of Black Insurgency Chicago ILUniversity of Chicago Press

mdashmdashmdash 1996 ldquoConceptual Origins CurrentProblems Future Directionsrdquo Pp 23ndash40 inComparative Perspectives on Social MovementsPolitical Opportunities Mobilizing Structuresand Cultural Framings edited by D McAdam JD McCarthy and M N Zald New YorkCambridge University Press

McAdam Doug and Yang Su 2002 ldquoThe War atHome Antiwar Protests and Congressional Voting1965 to 1973rdquo American Sociological Review67696ndash721

McAdam Doug Sidney Tarrow and Charles Tilly2001 Dynamics of Contention New YorkCambridge University Press

McCarthy John D Clark McPhail and Jackie Smith1996 ldquoImages of Protest Estimating SelectionBias in Media Coverage of WashingtonDemonstrations 1982ndash1991rdquo AmericanSociological Review 61478ndash99

McCarthy John D and Mayer N Zald 1977ldquoResource Mobilization and Social Movements A



Wed 28 Oct 2009 035926

Partial Theoryrdquo American Journal of Sociology821212ndash39

McVeigh Rory 2009 The Rise and Fall of the KuKlux Klan Minneapolis MN University ofMinnesota Press

Meyer David S and Debra Minkoff 2004ldquoConceptualizing Political Opportunityrdquo SocialForces 821457ndash92

Meyer David S and Sidney Tarrow 1998 The SocialMovement Society Contentious Politics for a NewCentury Lanham MD Rowman and Littlefield

Minkoff Debra C 1997 ldquoThe Sequencing of SocialMovementsrdquo American Sociological Review62779ndash99

mdashmdashmdash 2002 ldquoMacro-Organizational AnalysisrdquoPp 260ndash85 in Methods of Social MovementResearch edited by B Klandermans and SStaggenborg Minneapolis MN University ofMinnesota Press

Morris Aldon 1984 The Origins of the Civil RightsMovement New York Free Press

Myers Daniel J and Beth Schaefer Caniglia 2004ldquoAll the Rioting Thatrsquos Fit to Print SelectionEffects in National Newspaper Coverage of CivilDisorders 1968ndash1969rdquo American SociologicalReview 69519ndash43

Oliver Pamela E and Daniel J Myers 1999 ldquoHowEvents Enter the Public Sphere Conflict Locationand Sponsorship in Local Newspaper Coverage ofPublic Eventsrdquo American Journal of Sociology10538ndash87

Pierson Paul 2000 ldquoIncreasing Returns PathDependence and the Study of Politicsrdquo TheAmerican Political Science Review 94251ndash67

Piven Frances F and Richard A Cloward 1977Poor Peoplersquos Movements Why They SucceedHow They Fail New York Vintage

Polletta Francesca 2002 Freedom Is an EndlessMeeting Democracy in American SocialMovements Chicago IL University of ChicagoPress

Poole Keith T and Howard Rosenthal 2008Nominate Data Retrieved March 25 2008(httpwwwvoteviewcom)

Putnam Robert D 2000 Bowling Alone TheCollapse and Revival of American CommunityNew York Simon and Schuster

Ragin Charles C 1987 The Comparative MethodMoving Beyond Qualitative and QuantitativeStrategies Berkeley CA University of CaliforniaPress

mdashmdashmdash 2000 Fuzzy-Set Social Science ChicagoIL University of Chicago Press

mdashmdashmdash 2008 Redesigning Social Inquiry FuzzySets and Beyond Chicago IL University ofChicago Press

Ragin Charles C Kriss A Drass and Sean Davey2006 Fuzzy-SetQualitative Comparative Analysis20 Department of Sociology University ofArizona Tucson AZ

Rohlinger Deana A 2002 ldquoFraming the AbortionDebate Organizational Resources MediaStrategies and Movement-CountermovementDynamicsrdquo The Sociological Quarterly43479ndash507

Ryan Charlotte 1991 Prime Time Activism MediaStrategies for Grassroots Organizing Boston MASouth End Press

Sampson Robert J Heather MacIndoe DougMcAdam and Simon Weffer-Elizondo 2005ldquoCivil Society Reconsidered The Durable Natureand Community Structure of Collective CivicActionrdquo American Journal of Sociology111673ndash714

Schofer Evan and Marion Fourcade-Gourinchas2001 ldquoThe Structural Context of CivicEngagement Voluntary Association Membershipin Comparative Contextrdquo American SociologicalReview 66806ndash28

Schudson Michael 2002 ldquoThe News Media asPolitical Institutionsrdquo Annual Review of PoliticalScience 5249ndash69

Skocpol Theda 2003 Diminished Democracy FromMembership to Management in American CivicLife Norman OK University of Oklahoma Press

Snow David A Sarah A Soule and HanspeterKriesi eds 2004 The Blackwell Companion toSocial Movements Malden MA BlackwellPublishing

Soule Sarah A and Jennifer Earl 2005 ldquoAMovement Society Evaluated Collective Protestin the United States 1960ndash1986rdquo Mobilization10345ndash64

Tarrow Sidney 1994 Power in Movement NewYork Cambridge University Press

Tilly Charles 2005 ldquoRegimes and Contentionrdquo Pp423ndash40 in The Handbook of Political SociologyStates Civil Societies and Globalization editedby T Janoski R R Alford A M Hicks and MA Schwartz New York Cambridge UniversityPress

mdashmdashmdash Nd ldquoNorth American Social MovementOrganizations 1600ndash1999rdquo Columbia UniversityNew York

Vliegenthart Rens Dirk Oegema and BertKlandermans 2005 ldquoMedia Coverage andOrganizational Support in the DutchEnvironmental Movementrdquo Mobilization AnInternational Journal 10365ndash81

Wilson James Q 1973 Political Organizations NewYork Basic Books

Wooldridge Jeffrey M 2002 Econometric Analysisof Cross Section and Panel Data Cambridge MAMIT Press

Zald Mayer N and John D McCarthy 2002 ldquoTheResource Mobilization Research ProgramProgress Challenge and Transformationrdquo Pp147ndash71 in New Directions in ContemporarySociological Theory edited by J Berger and MZelditch Lanham MD Rowman and Littlefield



Wed 28 Oct 2009 035926

2005 review in Earl 2004) In this study weaddress why some SMO families receive exten-sive newspaper coverage by developing newdata on New York Times articles that mentionedUS SMOs across the twentieth century Wefirst identify which US SMOs and SMO fam-ilies have received the greatest newspaper cov-erage we then use this information tosystematically address why some movementfamilies receive extensive coverage appraisingwell-known theories of social movements andmovement consequences

Explaining newspapersrsquo SMO coverage isimportant for several reasons SMOs seek topromote many sorts of social change from cre-ating interests and identities to spurring politi-cal participation and civic engagement towinning political goals (Amenta 2006 Andrews2004 Clemens 1997 Gamson 1990 Ganz2000 McAdam 1982 McCarthy and Zald 1977Polletta 2002 Sampson et al 2005 Skocpol2003) and media coverage is important to theseefforts Coverage also constitutes key data inmapping political interests and identities amongthe politically disadvantaged it provides a meas-ure of discursive presence or influence in theproduction of culture akin to Gamsonrsquos (19901998) ldquoacceptancerdquo (Earl 2004) Using fuzzy setqualitative comparative analyses (fsQCA) across2153 movement family years we explore whysome movement families received extensivecoverage employing arguments from the dis-ruption perspective and the resource mobiliza-tion and political contextual theories We alsodevelop a relatively new political contextualargument enforced policies for a movementrsquosconstituency will spur movements and theircoverage

MOTIVATION PREVIOUS WORKAND MODELS OF MOVEMENTINFLUENCE

COVERAGE AS A CULTURAL CONSEQUENCE

OF MOVEMENTS

SMOs have been central to movement researchsince the early 1970s (Gamson 1990 McCarthyand Zald 1977) but few studies go beyondexamining one movement (cf Gamson 1990

Skocpol 2003) Moreover the mass news mediahave the widest gallery of all forums in the pol-icy-making process (Gamson 2004) so theattention SMOs receive in the mass media bol-sters their positions as representatives for theinterests and constituencies they claim (Ferreeet al 2002 Koopmans 2004) The mass mediahelp legitimize SMOs in a democratic politicalsystem in which most organized groups cangain some access to political institutions mediacoverage itself is a demonstration of SMOsrsquoimpact or acceptance (cf Gamson 1990) Manyalso see mass media coverage as necessary formovements to be influential (Lipsky 1968)SMOs seek to showcase and transmit their caus-es to relevant third parties and bystanders(Gamson 2004) by offering alternative framingsof issues (Cress and Snow 2000 Ferree et al2002 Ryan 1991) or discrediting opponentsand their framings (Gamson 2004) SMOs cangain coverage and influence policy debates inmultiple ways aside from protest (Amenta 2006Amenta Caren and Olasky 2005 Andrews2004) and those that receive coverage also tendto gain support (Vliegenthart et al 2005) Inshort media coverage of SMOs across move-ments and over time is an important if limitedconsequence of movements

SMOs appear in newspapers in differentways but always as a function of the practicesof newsgathering organizations which are con-cerned with generating ldquostoriesrdquo and ldquonewsrdquo(see Schudson 2002) Unlike with protest events(see review in Earl et al 2004) there is no wayto compare coverage of SMOs with all theirrelevant activity or all dimensions of their sizeIt is possible however to compare SMO cov-erage with important measures of movementsize such as membership and organizationaldensity and with protest events and other dis-ruptive activities Most important by compar-ing across all SMO families over a century wecan test theories about social movements andmovement consequences to explore why someSMO families achieved high coverage

FOUR THEORETICAL APPROACHES TO

EXPLAINING MOVEMENTS AND OUTCOMES

Prominent ideas in the literature on the conse-quences of social movements suggest first thatdisruption brings influence for movements Inthe classic view (Piven and Cloward 1977)


lective action a set of organizations subsumed by ourdefinition


Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926













Wed 28 Oct 2009 035926





Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926













Wed 28 Oct 2009 035926





Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926













Wed 28 Oct 2009 035926





Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926













Wed 28 Oct 2009 035926





Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926













Wed 28 Oct 2009 035926





Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926













Wed 28 Oct 2009 035926





Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926





Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926











Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926










Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926



















Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926





















equationsdisplay

equations

The extra







Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926




CONCLUSIONS









Dope DOpE DOp




Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926









Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926








REFERENCES

















Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926


































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926




































Wed 28 Oct 2009 035926

Documents

All the Movements Fit to Print: Who, What, When, Where ...oliver/SOC924/Articles/AmentaEtAl2009A… · the SMO family level, coverage correlates highly with common measures of the