Data at the Diocesan Level: Common Data Practices and Data at the Diocesan Level: Common Data Practices and Challenges Among U.S. Catholic School Superintendents Challenges Among U.S. Catholic School Superintendents

: Accountability pres sures in edu ca tion have risen steadily over the last two decades and pub lic schools and dis tricts now track school-and stu dent-level data in response to state and fed eral man dates. Cath o lic schools and dio ceses have not faced the same level of reg u la tion over this period, and less is known about data access and use in the Cath o lic sec tor. This descrip tive and explor atory research draws on sur vey and inter view data from a national sam ple of Cath o lic school super in ten-dents to exam ine data prac tices in dioc e san cen tral offices as well as bar ri ers faced in the use of data. Findings sug gest that although con sid er able var i a tion exists among dio ceses, com mon data efforts include work ing toward data cen tral i za tion, devel op ing tools to mon i tor the oper a tional health of schools, and find ing ways to fos ter the data cul ture within dio ceses. These prac tices were pres ent in close to half of all dio ceses, but sev eral com mon chal lenges related to gov er nance, resources, and data sys tems hin dered prog ress in and toward these efforts.

efforts have been funded by mul ti ple mil lion-dol lar fed eral grants (Coburn & Turner, 2012;National Center for Education Statistics, 2021), allowing pub lic schools to develop cen tral ized data sys tems at the state and dis trict lev els (Means et al., 2010).
As pri vate insti tu tions, Cath o lic schools have not expe ri enced the same level of reg u la tion or funding for data infra struc ture over this same period. Whether and how data prac tices have devel oped in Cath o lic schools is less clear. Despite being the larg est sys tem of pri vate schools in the United States (McDonald & Schultz, 2021), the orga ni za tion of the Cath o lic sys tem diff ers substan tially from the pub lic sys tem. Cath o lic school gov er nance is located at the school-rather than the dioc e san-(or dis trict-) level. As a result, Cath o lic schools are connected loosely in a dio cese and may face diff er ent chal lenges than pub lic schools in advanc ing data prac tices.
Understanding the data land scape in Cath o lic schools is impor tant for a num ber of rea sons. First, Cath o lic schools may not face reg u la tory pres sures, but as tuition-driven insti tu tions they do face mar ket pres sures and have expe ri enced decreased enroll ment over time (McDonald & Schultz, 2021). Second, oper a tions data along with achieve ment data are nec es sary for school lead ers to make deci sions at the school-and class room-level and for par ents as they choose schools for their chil dren to attend. Third, we lack a cur rent sense of how Cath o lic school stu dents com pare academ i cally to their pub lic school peers. Previous national stud ies com par ing aca demic achieve ment in Cath o lic and pub lic schools are now a decade or more old (see Bryk et al., 1993;Carbonaro & Covay, 2010;Coleman & Hoffer, 1987;Reardon et al., 2009). Additionally, with school choice funding avail able in a grow ing num ber of states, many Cath o lic schools are fac ing increas ing account abil ity pres sures from state and local agencies. Perhaps most impor tantly, data use is consid ered an impor tant strat egy for improv ing schools and edu ca tional offer ings (Coburn & Turner, 2012;Goldring & Berends, 2009), which could ben e fit schools in the Cath o lic sec tor.
This descrip tive and explor atory research draws on a frame work for data use by Coburn and Turner (2011) and seeks to under stand the orga ni za tional con text in which Cath o lic school cen tral offices are approaching data. Specifically, I con sider what types of data are avail able to Cath o lic school super in ten dents as well as what data prac tices they employ and the chal lenges they face in these efforts. A deeper under stand ing of the cur rent national con text may bet ter posi tion dioc e san and Cath o lic school lead ers as they build data infra struc ture, develop rou tines, and cre ate tools to assess school strengths as well as iden tify oppor tu ni ties for growth.

Literature Review
Greater account abil ity has increased the col lec tion of edu ca tional data includ ing stan dardized test scores, school-and stu dent-level demo graph ics, and teacher obser va tion data. Research sug gests that the value of these data is depen dent on their use (Coburn & Turner, 2011), and policymakers have touted data-informed deci sion-mak ing as a means for improv ing school-and stu dent-out comes (Goldring & Berends, 2009). Education research ers point to data use as crit i cal for under stand ing the link between edu ca tional data and school improve ment (Coburn & Turner, 2012;Grissom et al., 2017;Marsh, 2012;Spillane, 2012).

Framework for Data Use
Data use encompasses mul ti ple steps from data access to anal y sis to action (Grissom et al., 2017) that likely diff er school to school. Coburn and Turner's (2011) frame work for data use acknowl edges that under stand ing the orga ni za tional con text is a pre req ui site to design ing ways to encour age data prac tice as well as deter min ing the out comes related to data use. Within the orga niza tional con text of schools, Coburn and Turner argue that dis trict-and school-spe cific data routines draw atten tion to what data are wor thy of focus and dis cus sion. Data avail abil ity and access, how time and resources are applied to data activ i ties, the norms that shape inter ac tion with data, the role of lead er ship in data activ i ties, orga ni za tional deci sion-mak ing, and author ity struc tures all influ ence how schools inter act with data. Thus, an under stand ing of con text is nec es sary in order to plan for and enact suc cess ful orga ni za tional change in data prac tice (Coburn & Turner, 2011).
In edu ca tion, the orga ni za tional con text exists at mul ti ple lev els-dis trict/dio cese, school, and class room. To under stand data use, research ers have con sid ered it a sys tems prob lem, exam in ing the rela tion ships between the diff er ent lev els and the actions of each (Honig & Venkateswaran, 2012). Research sug gests the dis trict level plays an impor tant role in advanc ing data prac tice in schools and class rooms (Anderson et al., 2010;Datnow et al., 2007;Farrell, 2015;Honig & Venkateswaran, 2012;Park & Datnow, 2009;Wohlstetter et al., 2008). This rela tion ship is not one-sided, how ever. Central offices rely on access to school data while schools rely on cen tral offices for help in under stand ing and mak ing use of data (Honig & Venkateswaran, 2012).

Cath o lic Diocesan and School Context
The U.S. Cath o lic school and dioc e san con text diff ers con sid er ably from the more top-down U.S. pub lic school sys tem. Cath o lic schools are more accu rately described as a loose sys tem of schools rather than a school sys tem, grouped into 175 geo graph i cal dio ceses in the United States (McDonald & Schultz, 2021). While pub lic dis tricts can range in size and num ber of schools, there is greater var i a tion among Cath o lic dio ceses, which can range from a large met ro pol i tan area to an entire state. A sin gle dio cese can encom pass rural, urban, and sub ur ban schools. The larg est met ro pol i tan dio ceses have a hun dred or more schools in close prox im ity while the most rural dio ceses have fewer than 10 schools spread across hun dreds of miles. The major ity of Cath o lic dio ceses fall some where between the two (McDonald & Schultz, 2021). The num ber of per son nel in dioc e san school offices also varies from one part-time admin is tra tor in a small dio cese to more than 40 in the larg est dio ceses. Roughly one quar ter of dioc e san edu ca tion offices are staffed by one super in ten dent and one admin is tra tive sup port per son (McDonald & Schultz, 2021).
The dioc e san super in ten dent role also diff ers from its pub lic coun ter part. Whereas pub lic school super in ten dents make deci sions and insti tute ini tia tives and pol i cies for dis trict schools, the Cath o lic school super in ten dent serves in a more rela tional role and many act only in an advi sory capac ity (Brown, 2010). While there is var i a tion, a Cath o lic school super in ten dent's author ity often depends on the lead er ship style of the bishop and the his tor i cal rela tion ship between the dio cese and its schools (Brown, 2010).
Cath o lic schools are com prised of diff er ent school types, each with local con trol and own er ship. The major ity are par ish-owned and oper ated ele men tary schools that usu ally serve stu dents from pre school through eighth grade. With declin ing enroll ment, many par ish schools have closed or con sol i dated, becom ing jointly owned by mul ti ple parishes and governed by a board. Some par ish schools have been ceded to the dio cese for oper a tion and man age ment, and many dio ceses spon sor high schools. Other forms of gov er nance include inde pen dent schools owned and oper ated by reli gious com mu ni ties, the major ity of which are high schools, as well as newer sys tems or consor tiums of schools shar ing resources and board lead er ship (Goldschmidt & Walsh, 2013). While these inde pen dent and con sor tium schools serve stu dents with the bless ing of the bishop, the level of dioc e san involve ment varies con sid er ably.
The prin ci ple of sub sid i ar ity, man dated by Canon Law of the Cath o lic Church, allows for local con trol of Cath o lic schools, enabling pas tors and school lead ers to make deci sions according to the needs of the local com mu nity (Brown, 2010). This means, how ever, that the dioc e san super inten dent often has less say over what goes on in indi vid ual schools (Brown, 2010). As the sup port, involve ment, and direc tion varies, the rela tion ship between Cath o lic schools and the cen tral office is sub stan tially diff er ent from the pub lic sys tem.
Cath o lic schools have not expe ri enced the same pres sures to increase data use, but they also have not had access to the state and fed eral grants, often worth mul ti ple mil li ons of dol lars, that depart ments of edu ca tion have invested into cen tral ized, state wide data sys tems (Coburn & Turner, 2012). These sys tems and their related infra struc ture and use rou tines have taken mul ti ple years to build and develop (Means et al., 2010). As aca demic out comes of pub lic schools become more trans par ent, Cath o lic schools face increas ing mar ket pres sures related to stu dent-and schoollevel out comes.

Cath o lic Sector Data Needs
Multiple stud ies cite strong aca dem ics as a pri mary rea son that par ents choose Cath o lic schools (Convey, 1986;Lockwood, 2014;Lopez Arends, 2021;Trivitt & Wolf, 2011). Descriptive, crosssec tional com par i sons of stu dent achieve ment show that Cath o lic school stu dents score higher on stan dard ized tests such as the National Assessment of Educational Progress (NAEP; DeBrey et al., 2021) and NWEA's MAP Growth (Dallavis et al., 2021) com pared to their pub lic school peers.
However, addi tional lon gi tu di nal and matched ana ly ses are needed to bet ter account for selec tion bias-that there is some thing unique about stu dents who attend Cath o lic schools that is diffi cult to mea sure. The lack of com pa ra ble demo graphic infor ma tion and test data for Cath o lic schools and their stu dents hin ders such ana ly ses.
Religious edu ca tion and faith for ma tion are also impor tant for many par ents who choose Catho lic schools (Cohen-Zada & Sander, 2008;Lockwood, 2014;Lopez Arends, 2021;Sander, 2005). Cath o lic schools have placed greater empha sis on strength en ing the Cath o lic iden tity of schools in the tran si tion from vowed reli gious sis ters, broth ers, and priests to lay prin ci pals and teach ers in recent years (Convey, 2012;NSBECS, 2021). Resources such as the Assessment of Child/Youth Religious Education (ACRE) mea sure stu dent cat e chet i cal knowl edge and the National Standards and Benchmarks for Effective Cath o lic Elementary and Secondary Schools (NSBECS) help cat a log what schools do to fos ter faith for ma tion, mea sur ing Cath o lic iden tity and faith-related out comes con tin ues to pres ent chal lenges for school lead ers.
Because Cath o lic schools are pri vate and mostly tuition-driven, they face imme di ate con cerns of oper a tional health that are less of a con cern for indi vid ual pub lic schools. Over the last two decades, roughly 2,000 Cath o lic schools have closed (McDonald & Schultz, 2021) and issues of enroll ment, resources, and finan cial via bil ity are on prin ci pals' and dioc e san lead ers' minds. The deci sion to close a school impacts indi vid u als and fam i lies in both the par ish and local com mu nity (Brinig & Garnett, 2014), and accu rate data is needed to inform such deci sions and to iden tify where addi tional sup port might proactively avoid clo sure.

Research on Data Practices and Barriers to Use
Previous research on data prac tices in pub lic school sys tems can pro vide some insight into data prog ress in Cath o lic schools. In 2010, the U.S. Department of Education pro duced a report on data use fol low ing the sig nifi cant funding and invest ment in dis trict data sys tems with the pas sage of No Child Left Behind Act in 2002. While nearly all pub lic dis tricts had begun using stu dent infor ma tion sys tems (SIS), just over three-quar ters of dis tricts had stu dent assess ment sys tems or data ware houses to col lect and store lon gi tu di nal data. More than half reported using mul ti ple data sys tems to store and work with diff er ent forms of data (Means et al., 2010). Several stud ies have iden ti fied com mon bar ri ers to school data use, includ ing lack of time, skills or per son nel, and access to data (Grissom et al., 2017;Marsh, 2012). Others have iden ti fied issues with mul ti ple data sys tems that are not inte grated (Grissom et al., 2017;Means et al., 2010) as well as inad e quate pro fes sional devel op ment (Grissom et al., 2017;Means et al., 2010).
A review of the research (Honig & Venkateswaran, 2012) documented ways in which pub lic dis trict offices report addressing chal lenges and supporting school lead ers. These include assisting with data access by help ing to col lect, dis ag gre gate, and report data; help ing schools under stand data through goal set ting and bench marks, engag ing in crit i cal data con ver sa tions with prin ci pals; pro vid ing tools, resources, or time to sup port data use; devel op ing and insti tut ing expec ta tions for data use; offer ing train ing through pro fes sional devel op ment; and mod el ing the use of data at the dis trict level. Across these efforts, Honig and Venkateswaran (2012) argue that "rela tion ships mat ter" (p. 216) and high light the mutual depen dence between dis trict and school. Other research sug gests cen tral office per son nel are key play ers in data efforts and deci sion mak ing at the school level (Anderson et al., 2010).
One study exam ined diff er ences between pub lic dis tricts and char ter man age ment orga ni zations and found that "struc ture and deci sion-mak ing rights, size and growth tra jec tory, finan cial resources, and degree of reg u la tion [that] restricted or facil i tated the sys tems' mobi li za tion of resources" with regard to data (Farrell, 2015, p. 439). Thus, it fol lows that with diff er ences in con text-gov er nance, struc ture, resources, and account abil ity pres sures-Cath o lic school data prac tices and chal lenges may mir ror pub lic schools in some respects, but look some what diff er ent in oth ers.

Data and Methods
To bet ter under stand data con text in Cath o lic schools, I began at the dioc e san level to unpack the embed ded lay ers of data use (Coburn & Turner, 2012) using a mixed meth ods approach of sur veys and inter views with Cath o lic school super in ten dents. Initially, a sur vey was dis trib uted at the super in ten dent meet ing at the National Cath o lic Educational Association (NCEA) annual con ven tion in 2019. The sur vey sought to under stand data access, col lec tion, and use at the dioc e san level. Following the meet ing, the sur vey was dis trib uted over email to the full national pop u la tion of 175 Cath o lic school super in ten dents. After the removal of dupli cate responses, the ini tial sur vey had a 34% response rate. Survey data was com bined with data from the inter view sam ple (described in sub se quent par a graphs) and together com prised a 39% response rate (n = 68). Although the sam ple can not be con sid ered fully rep re sen ta tive, all regions and dioc e san sizes were pres ent in the data (see Table 1).
To iden tify the inter view sam ple, I ran domly drew from the entire pop u la tion of 175 Cath o lic dio ceses and arch di o ceses, 1 strat i fy ing the sam ple first by num ber of schools in the dio cese and sec ond by region, with an oversample of the larg est dio ceses, using the NCEA Annual Statistical Report on Schools, Enrollment, and Staffing (McDonald & Schultz, 2019; see Table 1). The oversample pro vided more than one per spec tive for the larg est dio ceses. I emailed invi ta tions to par tic i pate to the super in ten dent and replaced dio ceses that did not respond or declined using a ran dom draw of the same size and region.
I conducted inter views between March 2020 and Novem ber 2020 over Zoom with super in tendents or their del e gates, usu ally an assis tant or asso ci ate super in ten dent. Two inter views included more than one indi vid ual. Interviews lasted between 45-60 min utes and were audio-recorded. Using a semi-struc tured pro to col, I first gained a sense of the size and con text of the dio cese before mov ing to a set of com mon ques tions regard ing dioc e san pri or i ties, data prac tices, per ceived data chal lenges, and aspi ra tions for data use.
Relevant sur vey data are presented within the text using descrip tive sta tis tics. Interview data were tran scribed for anal y sis and coded using an iter a tive pro cess to sur face com mon themes and pat terns. Early read ings of the data focused on data access, prac tices or rou tines, and chal lenges, looking for areas where super in ten dents report investing the most effort into data. In sub se quent read ings, broad categories were refined, looking inten tion ally for disconfirming evi dence and out ly ing cases to gain a sense of var i a tion (Saldaña, 2012). Finally, I mapped codes by num ber of schools and geo graphic region to con sider the pos si bil ity of pat terns in responses based on dioc esan char ac ter is tics.

Findings
Three com mon efforts-work ing toward data cen tral i za tion, cre at ing tools to mon i tor oper ational health, and grow ing the data cul ture-were described by super in ten dents in the inter views. I found that a con sid er able share-roughly half-of the dio ceses had made sub stan tial prog ress in these efforts while other dio ceses were begin ning or con sid er ing this work. Survey and inter view data com ple ment each other. I pres ent rel e vant sur vey data first, followed by inter view data to pro vide fur ther con text.

Working Toward Data Centralization
Cath o lic dio ceses have been on their own to pur sue data cen tral i za tion, often with few resources and lim ited staffing. Some dio ceses have made sub stan tial prog ress by adopting com mon stu dent infor ma tion sys tems (SIS) and com mon interim or annual assess ments in all schools in the dio cese. The use of uni form sys tems with cen tral office access allows super in ten dents the abil ity to mon i tor trends in the data, with out hav ing to col lect infor ma tion from each school, a timely endeavor for dioc e san and school per son nel.
Survey results suggested that in more than one third (38%) of dio ceses, all schools use the same SIS (see Table 2). In 28% of dio ceses, there are two SIS in use. In another third (32%), there are three or more sys tems in use across diff er ent schools, with a few dio ceses reporting as many as six sys tems. Only one dio cese in the sur vey reported not using an SIS. Interview data suggested that in some dio ceses, P-8 schools use a com mon sys tem while high schools use a diff er ent sys tem.
Roughly 37% of dio ceses sur veyed reported that all P-8 schools are using the same interim assess ment (see Table 2). Thirty-one per cent report two tests and 25% report using three or more interim assess ments across the dio cese, with a few dio ceses reporting as many as five. Roughly 7% of dio ceses did not include an interim assess ment in the sur vey. Some dio ceses with out interim assess ments con tinue to use annual assess ments. Close to half of the dio ceses sur veyed report using one annual assess ment. Thirty-five per cent report tak ing a state assess ment. Interview data con firms the trend that schools and dio ceses increas ingly are adopting interim assess ments for mon i tor ing stu dent achieve ment mul ti ple times dur ing the school year.
In the inter views, I found that cen tral iz ing and stan dard iz ing data is on the mind of most super in ten dents. Some have made this orga ni za tional shift while oth ers are work ing toward it. Of the 27 super in ten dents interviewed, over half had a com mon SIS and more than three-quar ters had some form of com mon assess ment. Just over half had both a com mon SIS and assess ment with cen tral access. Dioceses with these sys tems were pres ent in all regions and size categories. Several super in ten dents also discussed plans for com mon finan cial sys tems.
Having cen tral ized access to schools' data allows the super in ten dent to pro vide addi tional support to schools and to diag nose prob lem areas. As one super in ten dent stated, "Our schools can't do it by them selves . . . . I think by get ting more data we can plan on how to help with enroll ment man age ment, how to mar ket, how to advise prin ci pals." With addi tional data, super in ten dents can pro vide tai lored assis tance based on spe cific needs. Another super in ten dent described the desire for data to help schools work more closely together: "I want to get every body on the same sys tem so that we can col lect data together and we can use that data to drive our goals." Using cen tral ized data can help bring schools together to address com mon areas in need of atten tion.
Superintendents are also seek ing greater stan dard i za tion, espe cially related to tuition and sal ary prac tices. Without cen tral ized data or data shar ing among schools, super in ten dents have lim ited access to this infor ma tion: Up until this year, the trans par ency wasn't there for me to see what the other schools are charg ing for tuition . . . . Those types of data points are impor tant for us to dis cuss because I want to make sure that all of our schools are com pet i tive.
For most super in ten dents this cen tral i za tion and stan dard i za tion is a work in prog ress, but many rec og nized the pos si ble advan tages that could come with increased coop er a tion among local schools. Particularly, greater num bers would allow schools to take advan tage of effi cien cies and econ o mies of scale, as sys tems and assess ments can be expen sive for a sin gle school to adopt on its own. As one more cen tral ized dio cese explained, When we feel like it is a dioc e san-wide ini tia tive, we do it in all [of our] schools. There are prob a bly few things we would pilot. We try to have it all laid out before hand for effi cien cies and econ o mies of scale. This nego ti at ing for group and larger scale pric ing came up in sev eral inter views for those who are cur rently using com mon sys tems, ref er enc ing an option that is sim ply not avail able to indi vid ual schools.
In addi tion to cost effec tive ness, cen tral office staff rec og nize that these sys tems can pro vide increased effi ciency for school and cen tral office per son nel. As one dio cese that recently made the change described the pro cess: There was a lit tle bit of pushback, but now it's like, why weren't we doing this for years?... It's just more effi cient and more effec tive . . . . Everyone's on board now and it's taken us a while to get there, but I'm very proud and they're proud. And it light ens the load a lit tle bit. Gives access quicker.
Several super in ten dents discussed the time-sav ing nature of hav ing cen tral ized infor ma tion and the abil ity to down load needed infor ma tion from the sys tem with out hav ing to col lect it. As another super in ten dent described it: If noth ing else, it just cuts down on the time because we get the ran dom, "how many kids do you have in after school care, " or, "what is the per cent age of this or that?" . . . Instead of hav ing to send an email to all the prin ci pals and then col lect the data then put it some where else and send it back to some one, we can just log in and pull it our selves right then . . . . It's always the most up-to-date infor ma tion with out the mid dle man.
Although those on cen tral ized sys tems are enthu si as tic about effi cien cies, they acknowl edge that get ting on uni form sys tems is not easy.
Because of local gov er nance, many super in ten dents are only able to encour age, rather than man date, a shift to uni form sys tems. Thus, some super in ten dents have begun lay ing ground work for sys tems months and years in advance, mak ing the case and pre par ing schools for change. An admit tedly slow pro cess, sev eral super in ten dents reported tak ing this grad ual approach. One super in ten dent discussed the move to a com mon sys tem as sim ply encour aged. Others report being "trans par ent" or "stra te gic" in bring ing up these ideas with school lead ers: I've found that if there's a big new ini tia tive, like a new stu dent infor ma tion sys tem, I start talking about it months before I need to get peo ple used to the idea. They can start complaining a lit tle bit and then I keep push ing a lit tle bit, but then I also found that if I use peo ple from their schools as part of the team they feel like they've been kept informed and they are a part of the jour ney.
School lead ers often require con vinc ing to make the move to data cen tral i za tion. Some dio ceses have offered incen tives that can assist in funding the tran si tion, tying use of sys tems to aid. This pro vi des what one super in ten dent described as "lever age." Another described a suc cess ful tran si tion as begin ning with pas tors: The pro cess started by talking to the pas tors of those schools that did not have it and say ing we have to get the money in here to do it and then work ing with [the com pa ny] to give us a dioc e san price, kind of a cut or ben e fit. The schools that we finally got on [the sys tem] . . . the dio cese had to help throw some money in and then of course we had to get the train ing done for their staff.
Similarly, some have linked dioc e san tuition and finan cial assis tance to the use of com mon sys tems. As one described it: "In order for each school to be eli gi ble for tuition assis tance, they had to get on board with the [fi nan cial] man age ment sys tem." Superintendents also shared that use of these sys tems, while help ful, was not always a pan a cea. One super in ten dent stressed the need for every one to use the sys tems: "You can have all the tools in the world but if you don't use them cor rectly or if peo ple don't feel com fort able to use those tools, then, you know, it's just a rock sit ting in the cor ner." For sev eral super in ten dents, suc cess ful use at the dioc e san level depends on accu rate and com plete infor ma tion entered at the school level. As one super in ten dent explained, "It takes a cer tain level of dil i gence in terms of vet ting and stay ing on top of that infor ma tion and reviewing it and push ing it back to the prin ci pals when it's incom plete." In addi tion, some super in ten dents expressed frus tra tion that many sys tems do not "talk" directly to each other or are not eas ily inte grated, and the mul ti ple sys tems-stu dent informa tion, test assess ments, finan cial sys tems, human resources-can be diffi cult for small offices to nav i gate: Because our data is not linked, that makes it a chal lenge. And that's one of the things that I found at the dioc e san level is, if you don't have a big office and you don't have soft ware. . . . They're usu ally on sep a rate data bases, sep a rate sys tems, that don't talk.
This causes frus tra tion because these indi vid ual sys tems require con sid er able effort to com bine data from mul ti ple sys tems.
Although many dio ceses have made the move to cen tral ized sys tems and oth ers view this as a pri or ity or in prog ress, there are some dio ceses that cur rently do not have plans to cen tral ize. As one described their sit u a tion: The way we're struc tured in our dio cese [is that] we are a sys tem of schools, not a school sys tem. So, each of our schools, quite frankly, is an inde pen dent entity. So, they have all of their data . . . . Some use it very well. Some not at all . But there is no way that I can use that data. I have no way to access it, nor would I be able to manip u late it in any way . . . . So, there's just no struc ture in place to really [centrally] use data well.
Unless the super in ten dent has the author ity to man date change or the will ing ness to find cre a tive and some times finan cial solu tions nec es sary to con vince schools to adopt these sys tems, some see lit tle pros pect for suc cess.
While gov er nance in Cath o lic dio ceses appears to be a sub stan tial chal lenge, in addi tion to cost and indi vid ual school-will ing ness, some super in ten dents' appe tite for change also deter mines the rate of prog ress toward cen tral i za tion. Considered against the Coburn and Turner frame work (2011), the move toward uni form sys tems is an effort to address the lack of access to school-level data in the cen tral office. Both the rela tion ship between the dio cese and its schools, and the superin ten dents' con cep tion of the role and its lim its appear to influ ence a dio cese's move toward data prog ress and the extent of data prac tice.

Developing Tools to Monitor Operational Vitality
Dioceses with and with out cen tral ized data, and across size and region, reported cre at ing tools to mon i tor the oper a tional health of Cath o lic schools. While these have mul ti ple names such as snap shot, score card, index, matrix, report, dash board, they are usu ally static sum ma ries that contain some com bi na tion of met rics on var i ous aspects of school oper a tion. Of the dio ceses sur veyed, 44% reported using a data mon i tor ing instru ment. Closer to 40% of dio ceses in the inter view sam ple discussed cur rent sum mary reports, and slightly less than a quar ter reported plans for devel op ing such reports.
Superintendents discussed using these tools to iden tify "gaps" to deter mine where to pro vide sup port. As one super in ten dent described, "I need to be like a coun try doc tor. I need to know where the ail ments are." For many super in ten dents, the focus of these instru ments is school oper a tional health. Another super in ten dent explained: "We can't know which schools are financially sound with out pulling in finan cial data or enroll ment data, reten tion, admis sions, inquiry, aca demic excel lence." Superintendents col lec tively suggested that a bet ter under stand ing of each indi vid ual school pro vi des insight into what the school sys tem looks like as a whole.
Several super in ten dents discussed the effort, "to serve the prin ci pals and the pas tors to help them run their schools as best they can, " as the rea son their role exists. Further, super in ten dents expressed hope that data will be used proactively through these tools. Some acknowl edged the ten dency toward more his tor i cal and reac tive data use in Cath o lic schools. One super in ten dent elab o rated: We're try ing really hard to look at the sus tain abil ity of our schools . . . What will our numbers look like ten years from now?... We're try ing to say to our com mu ni ties, "Hey, this is what we're antic i pat ing so let us help you plan for your future." Acknowledging that pas tors and school lead ers may not be busi ness and mar ket ing experts, and that some may have vary ing exper tise as instruc tional lead ers, super in ten dents seek to empower stake hold ers with data and orga ni za tional tools needed to make informed deci sions. To do so, super in ten dents described collecting a range of data points, "any thing that could come up that could cause trou ble or that can high light the pos i tives." As another put it, "It's finan cial. It's aca demic. It's oper a tional. It's enroll ment. It's demo graphic. And it's just really pulling all of that into some kind of coher ent fash ion to make sense." These mon i tor ing reports often include some form of bench mark ing, as one super in ten dent explained: "We cre ated a rubric around aca dem ics, finance, enroll ment, lead er ship and some other areas to pro vide a data snap shot of our schools that will allow us to pro vide the inter ven tions or deter mine what are our most crit i cal schools." Superintendents reported devel op ing these snap shots over time, first by iden ti fy ing the nec essary met rics, followed by data col lec tion or orga ni za tion, and often as part of a multi-year pro ject: "We started looking at the var i ous met rics of school per for mance. So, every thing from oper a tional vital ity to aca demic excel lence. And we tried to say, what are the right mea sures? And we're constantly refin ing those." One super in ten dent described their dio cese's prog ress as "still kind of in a vision ing phase of it. So, I have to fig ure out what, you know, when we look at a pic ture of a school, what are those data points?" Another shared their prog ress: The dash board prob a bly won't come for another two or three years. The first year or two is going to be let's get these assess ments in place. Let's get this data col lected on a reg u lar basis and start reviewing it, see ing what it tells us and then once we get that, let's build that dash board.
In order to cre ate these tools, the cen tral offices must have data. Dioceses with cen tral access to school infor ma tion have an advan tage, but each of the sys tems-stu dent infor ma tion, assess ment, or finance sys tem-often require mul ti ple down loads, data orga ni za tion, and entry of infor ma tion into some form of report. As one super in ten dent described it: "The prob lem is we don't have soft ware that's going to auto mat i cally pop u late. I've got to do it by hand. It'll be a static snap shot and if I get it once a year, I'm lucky." These snap shots remain time-inten sive to cre ate. I found that the lack of cen tral access has not prevented some dio ceses from cre at ing these tools. Several dio ceses have launched sur veys and other data col lec tion ini tia tives to amass the data needed to cre ate these indi vid u al ized reports. Others have mined infor ma tion already col lected: So, we already have that infor ma tion, it was in a file. . . . We had to have a lot of staff meetings to fig ure out where is this stuff ?...We needed the capac ity to be able to mine it as we went along. . . . I would love for it to be a click of a but ton and do it. But it's still a lit tle labor inten sive.
While larger dio ceses often have ded i cated data per son nel, data tasks are the respon si bil ity of the super in ten dent in most small and medium-sized dio ceses. Some small dio ceses have iden ti fied tal ented teach ers that assist on a part-time basis. Regardless of who is doing the data work, there is an invest ment of time that some dio ceses are bet ter posi tioned to ded i cate to these instru ments than oth ers.
Superintendents described using these reports to ini ti ate con ver sa tion with school stake hold ers and lead ers. Several men tioned bring ing the report to the school leader and pas tor to begin dis cussion about oper a tional health. As one super in ten dent recounted: We can use [the mon i tor ing tool] to have just an ini tial con ver sa tion about a school's via bility and we're hav ing those con ver sa tions a bit more than we used to given the cur rent real ity that we're in. And again, it's a starting point [for con sid er ing inter ven tions] and [to] be able to hone in on what they need to do.
Superintendents described these meet ings or calls as oppor tu ni ties to "walk [school lead ers] through what's impor tant, " help ing to focus atten tion on iden ti fied areas of con cern. Some superin ten dents can be more direc tive but oth ers are lim ited to presenting the data, as one shared: "I might say to the prin ci pal, 'Hey, I'm looking at this and it tells me XYZ, you know, take it for what it's worth.'" Regardless, super in ten dents reported that a mon i tor ing tool "allows you to have some really good con ver sa tions with schools." These con ver sa tions enable super in ten dents to help school lead ers make sense of the data with the goal of for mu lat ing pro ac tive responses.
Nearly half of the dio ceses sur veyed and interviewed reported the use of some form of sum mary tool to cap ture data on diff er ent aspects of school oper a tion in order to mon i tor prog ress and iden tify areas of con cern. This prac tice requires not only access to data, shared either through cen tral access to com mon sys tems or through data col lec tion, but also the invest ment of resources and per son nel, namely in the time it takes to gather, orga nize, ana lyze, and pres ent the data to school stake hold ers. These mon i tor ing tools, when con sid ered against the Coburn and Turner (2012) frame work, draw atten tion to the needs of the schools and dio cese and high light the types of data of pri mary impor tance to the sec tor. The rou tines of gath er ing nec es sary data to gain a com pre hen sive under stand ing of school health and the com mu nal account abil ity related to the use of these sum mary reports pro vide a means for orga ni za tional change through pro ac tive assess ment of areas of vul ner a bil ity.

Growing Data Culture
Superintendents in all size dio ceses and regions acknowl edged and discussed efforts related to devel op ing a data cul ture among school lead ers. For some, this began with rec og niz ing the uncertainty involved in tak ing a close look at pro cesses and out comes. As one super in ten dent stated, "There's going to be a lot of data that might be scary to peo ple. How do we help them deal with that?" For some, this requires antic i pat ing responses and pro vid ing reas sur ance: I think a lot of the hes i ta tion [for the dash board] came from fear. I think the fact that noth ing bad hap pened, it was just a con ver sa tion that could help. We aren't try ing to penalize you; we are just try ing to help, and we needed some thing in our office [to look] across [schools].
Superintendents discussed using the data con struc tively, work ing with schools to address what the data might reveal. Another super in ten dent expressed this sim i larly, "We say, 'Okay. This is where we are, but this is where we are going to be and this is how we are going to get it right.'" Superintendents see these meet ings as steps for ward in addressing needs. Improved use of data can lead to changes in school pro cesses that can result in dis com fort. As one super in ten dent explained, "Change takes time. So, just get ting every one to the same cal i ber of con ver sa tion about data and really hav ing mean ing ful con ver sa tions about stu dents using the sup port ive data points that you know they have." Another super in ten dent saw the use of data as a way to help school lead ers embrace change: One thing that I think we could do with data is become more com fort able with the idea of change. and . . . instead of feel ing like we're a step behind or a beat late, it would bring us closer to the cut ting edge. I don't think that there's a school in the world that's not con stantly chang ing. Just by vir tue of our demo graph ics we're chang ing . . . . We've got to become more com fort able with that over time.
Here, data is described as a cat a lyst for change. To bet ter deal with change, super in ten dents discussed fur ther devel op ing rela tion ships and trust around the use of data, with the goal of "build ing col lab o ra tion and col le gi al ity." Superintendents reported that this requires rap port and trans par ency, a sense of con nect ed ness, and a reas sur ance that super in ten dent and prin ci pals share in this effort.
The more peo ple get to know us, and we get to know them and [they] trust us, I think it's eas ier to get them used to the idea that we're going to become a lit tle more cen tral ized, still while leav ing them auton omy as prin ci pal. I think that's prob a bly what they're wor ried about, but I'm not inter ested in tell ing them how to run their school. I just want to be able to col lect some data that will help us run the dio cese bet ter.
Relational trust is described as crit i cal: "it's build ing that trust and that desire in the leader to actually want to come along with us in this work." Some super in ten dents seek to develop trust through mod el ing of data prac tice. A few super in ten dents viewed their data efforts as help ing schools to see what is pos si ble with data. As one stated: If I'm going to make a diff er ence and if I'm going to influ ence peo ple, and I'm going to bring them along and help them see a bet ter way for ward that isn't going to kill them, but it's actu ally going to give them life, I have to model that, first of all . I have to prac tice what I preach.
By mod el ing data prac tice, super in ten dents develop expec ta tions and norms related to sys tem use and data col lec tion. These efforts often start from the ground up. As one super in ten dent explained: "I didn't real ize the pre req ui sites [school lead ers] didn't have with work ing with data and all of that. It just wasn't a part of their cul ture." Some super in ten dents see the need to build data cul ture within the dio cese, rec og niz ing the need to con vince some school lead ers of the uses and value of data, hoping to instill "a com bi na tion of the con fi dence and the exper tise to actu ally use data of any kind." Some super in ten dents discussed accessing the sense of com mu nity as a resource within the dio cese. As one super in ten dent stated: "In this dio cese, one of the things we try to do is say we're all in this together, we all have to help one another." By encour ag ing the shar ing of best prac tice, schools can ben e fit from each other's expe ri ences. Another super in ten dent discussed "find ing ways to . . . use our col lec tive power to address indi vid ual needs. So, if ...we're see ing a small trend across a few schools, how can we kind of con nect those together to focus on that area." Similarly, one super in ten dent described connecting two schools based on data-one school looking for a new math cur ric u lum with another that had recently made strong gains in math-to take advan tage of cur rent strengths to sup port antic i pated change. Another described how two schools might share enroll ment ideas and strat e gies: What I might say is, "This is a school that has a nearly iden ti cal demo graphic pro file as your school does . . . Their school is at 95% capac ity, yours is at 75% capac ity. So, maybe they're doing some thing diff er ent, and then maybe you can work with that prin ci pal to say, what is that some thing?" Identifying areas of strength and growth within a dio cese can assist in addressing school-spe cific issues.
Some super in ten dents, how ever, acknowl edged ret i cence on the part of schools to share with other schools, "We don't seem to read ily share out of fear, or I don't know com pe ti tion, or we just don't think to do it because we're so into what we're doing." Some see the shar ing of exper tise and expe ri ence as the way to move Cath o lic schools for ward: "The more we talk, the more we are brought together as Cath o lic lead ers, the more we share ideas and stuff. I think we have an opportu nity to really impact this min is try and to grow it again." Convincing school lead ers of the value of shar ing exper tise is an impor tant aspect of grow ing data cul ture.
Overall, cen tral office per son nel rec og nized the rela tional inroads that are needed for the adoption and use of data infor ma tion sys tems. This work involves build ing the trust of school lead ers in the col lec tion and use of data. Superintendents might also need to con vince school lead ers of the ben e fits of investing time and per son nel in data work as well as the advan tages of shar ing data. They must acknowl edge strong norms of local con trol, dis tinctly diff er ent bal ances of power, and ten sions related to com pe ti tion and fear of clo sure as they seek to advance data prac tices in the dio cese.

Discussion
Survey and inter view data suggested that Cath o lic dioc e san super in ten dents are mak ing some prog ress toward cen tral iz ing data and devel op ing mon i tor ing tools to assess oper a tional wellbeing. Superintendents also appeared to under stand the need to strengthen the data cul ture within their dio ceses. For those with cen tral ized access to stu dent infor ma tion and test scores, and even those who man u ally col lect data from indi vid ual schools, the cre a tion and use of mon i tor ing tools for pro ac tive con ver sa tions reflect an under stand ing of data use through steps of access, anal y sis, and action (Grissom et al., 2017). However, my find ings sug gest that while some dio ceses are using some form of sum mary instru ment, many dio ceses are not. Considerable var i a tion exists in how much data super in ten dents have access to, which influ ences the extent they are able to engage with data.
Evidence from this study ech oes pre vi ous find ings in pub lic schools related to how cen tral offices inter act with schools regard ing data. Similar to dis trict super in ten dents, dioc e san superin ten dents assist with data tasks for schools, help schools make sense of data, develop norms and expec ta tions through con ver sa tions, and model data use through mon i tor ing tool ini tia tives (Honig & Venkateswaran, 2012). They also report sim i lar chal lenges, includ ing time con straints, lim ited access to data, insuffi cient per son nel and exper tise (Grissom et al., 2017;Marsh, 2012), as well as sys tems that are not eas ily inte grated (Means et al., 2010). Public dis tricts reported these as chal lenges close to a decade ago, yet it is clear that Cath o lic dio ceses are just begin ning to grap ple with these issues. The data efforts described here-par tic u larly toward cen tral i za tion and mon i tor ing-sug gest sig nifi cant and pos i tive for ward prog ress. However, some dio ceses with out data sys tems and prac tices saw devel op ment as tak ing years, suggesting a lack of urgency, which some Cath o lic schools may not be able to afford.
Cath o lic dio ceses face addi tional chal lenges regard ing data as well. Without access to pub lic funding for data infra struc ture, they have fewer resources to devote to data sys tems and mon i tor ing reports. As the major ity of schools are tuition-depen dent, Cath o lic schools face diff er ent pres sures related to enroll ment and oper a tions than pub lic schools. I find that dioc e san super in ten dents have a slightly diff er ent focus in their data work. While oper a tional health of the school sys tem may only be of some con cern to pub lic school super in ten dents, many dioc e san super in ten dents pri ori tize the oper a tional vital ity of indi vid ual schools. In the Cath o lic school sys tem, aca demic data are just one data source among many, with enroll ment and other oper a tional aspects of run ning schools of equal or greater impor tance.
The gov er nance of Cath o lic schools also pres ents data chal lenges. In many cases, the decentralized, local con trol of schools does not allow super in ten dents to man date the use or adop tion of data sys tems. Instead, they must bring school lead ers to con sen sus, build rela tion ships and trust, and pro vide nec es sary ratio nale and moti va tion to move toward cen tral i za tion; a much slower and timecon sum ing pro cess that requires strong lead er ship and a will ing ness to engage. While not the focus of this paper, dioc e san super in ten dents are also strug gling with how to mea sure a sig nifi cant aspect of their mis sion: the Cath o lic iden tity of schools and the faith for ma tion pro vided to stu dents.
The con text of Cath o lic schools, how ever, does pro vide cer tain oppor tu ni ties for data use and prog ress. Within Cath o lic dio ceses, schools share a com mon mis sion of edu cat ing the whole child, of pro vid ing faith for ma tion and aca demic excel lence (Bryk et al., 1993). This shared mis sion may pro vide the starting point for dio ceses to build the foun da tion for increased data use. Cath o lic schools and dio ceses are described as hav ing a strong sense of com mu nity (Bryk et al., 1993;Coleman & Hoffer, 1987). Depending on the extent that dioc e san per son nel are able to cap i tal ize on this, a shared sense of com mu nity may enable the super in ten dent to rally school lead ers around the use of data sys tems and pro cesses for the com mon good of the school and dioc e san com mu nity. While local gov er nance pro vi des some chal lenges, Cath o lic schools pos sess the abil ity to use data at the local level to make school-spe cific deci sions and changes. The shar ing of suc cess ful data practice across dio ceses and schools may also help advance these efforts. Once chal lenges around data sys tems and anal y sis are addressed, Cath o lic schools may be bet ter posi tioned to pri or i tize local deci sions and act directly on infor ma tion in ways that best fit the needs of the local com mu nity.
Examining the dis tinct con text of data in the Cath o lic sec tor extends the under stand ing of data use in edu ca tion and applies pre vi ous research in pub lic dis tricts to pri vate and faith-based schools. Findings unique to the Cath o lic sec tor-the need for cen tral ized data and the devel op ment of tools to mon i tor oper a tional vital ity-point to the impor tance of con sid er ing data prac tice through a frame work such as Coburn and Turner's (2011) that acknowl edges the con sid er able var i a tion pres ent among school orga ni za tions, both within and across sec tor. Issues of gov er nance, norms, rela tion ships, lead er ship, and data access in the Cath o lic sec tor inform the inter ven tions that attempt to effect orga ni za tional change. While there are com mon prac tices, the local con text con tin ues to play a sig nifi cant role in how these evolve.
Although a national and rep re sen ta tive sam ple was selected at ran dom for the inter view, the dio ceses choos ing to par tic i pate may reflect the expe ri ences of more data-for ward dio ceses and thus may over state the use of data in Cath o lic schools and dio ceses. In addi tion, the sur vey, while includ ing dio ceses of all sizes and regions, gath ered data from just under 40% of the full pop u la tion of U.S. Cath o lic dio ceses and thus may not be fully rep re sen ta tive. This study drew on both sur vey and inter view data to ana lyze dioc e san per spec tives of cur rent data prac tice, but these meth ods did not pro vide a first hand view of cur rent data rou tines in action, nor did they pro vide insight to data rou tines at other lev els of the edu ca tion sys tem. To gain a full sense of data prac tices in the Cath o lic sec tor, future study would ben e fit from the exam i na tion of data prac tice in action at the dioc e san, school, and class room lev els.
Successful orga ni za tion, access, and use of data are but a few of the chal lenges fac ing Cath o lic schools and cen tral offices. An increased use of data and a clearer under stand ing of the cur rent state of schools-includ ing areas of vul ner a bil ity and strength as well as oppor tu ni ties for growth-may bet ter posi tion Cath o lic schools to suc ceed in an edu ca tional land scape of increased choice. Proactively exam in ing areas of vul ner a bil ity as well as oppor tu ni ties for growth can help Cath o lic schools to ful fill their mis sion to pro vide excel lent faith and aca demic for ma tion to the stu dents they serve.