Coming Soon to a Neighborhood Near You?


Off-Campus Recruiting by Public Research Universities

Karina Salazar

University of Arizona

Ozan Jaquette

University of California, Los Angeles

Crystal Han

University of California, Los Angeles

University of Arizona University of California, Los Angeles

ozanj.github.io/naed_presentation

Acknowledgments



This research was made possible by funding from the following sources:


  • National Academy of Education/Spencer Foundation (Postdoctoral Fellowship)
  • American Educational Research Association (Dissertation Grant)
  • UCLA Office of Equity, Diversity, and Inclusion (Faculty Career Development Award)
  • Joyce Foundation (Research Grant)

Motivation


The problem with policy discourse about college access

The 2014 White House "Access Summit"

  • The White House (2014a) review of causes of unequal college access
    • Highlights "achievement gap", "under-matching"
  • Commitments to Action on College Opportunity (The White House, 2014b)
    • Universities pledge "action plans" (e.g., holistic admission, need-based aid, "outreach")

Problem with policy discourse

  • Place responsibility on students, K-12 schools; assume universities are passive or progressive


Alternative explanation for access inequality

  • University enrollment priorities biased against poor students and/or communities of color

Research focus (this paper)

  • We argue recruiting behavior is an indicator of enrollment preferences
  • Research question: What are the similarities and differences in off-campus recruiting patterns across public research universities?

Background


The enrollment management industry

The enrollment funnel

Enrollment Funnel

Interventions along the funnel

  • Identify prospects
    • Purchase "student lists"
  • Recruit prospects remotely
    • Email, mail, text, etc.
  • Recruit prospects in-person
    • Off-campus recruiting visits (e.g., high school visits, fairs)
    • Campus visits
    • Other "outreach"
  • Solicit inquiries, stealth applicants
    • Social media, advertising
  • Convert admits to enrollees
    • Financial aid leveraging

Literature review


Scholarship on the enrollment funnel

Most research analyzes admissions (e.g., Killgore, 2009) or financial aid (e.g., McPherson, Schapiro, 1998)

  • Final stages of enrollment funnel

Scholarship on recruiting is sparse

  • Audits of university response to inquiries (e.g., Hanson, 2017; Thornhill, forthcoming)
  • Off-campus recruiting visits
    • From the college perspective (Stevens, 2007)
      • Important for relationships with prospects, counselors at "feeder" schools
    • From the perspective of high school students (Holland, 2019)
      • Which universities visit affects student decisions; especially first-gen, students of color


Research gap

  • We don't know which universities visit which schools and communities
  • If poor students, communities of color are not being recruited, then "under-matching" may be due to under-recruiting rather than lack of guidance

Theoretical motivation


Enrollment priorities and recruiting behavior

Organizational theory

  • Contingency theory (Thompson, 1967)
    • Technical level vs. managerial level
  • "New" institutional theory (Meyer and Rowan, 1977)
    • Publicly adopt goals demanded by environment
    • Technical level cannot pursue all goals
      • Substantively adopt some goals (technical level)
      • Symbolically adopt others (policies, rhetoric)


Application to enrollment management (EM)

  • "Iron triangle" of EM highlights three broad enrollment goals: access, academic profile, revenue
    • Resources scarce; depending on priorities, some goals receive more resources than others
  • Enrollment priorities cannot be discerned by policies, rhetoric
  • Recruiting is allocation of resources from technical level
    • Knowing which populations targeted by recruiting interventions indicate enrollment priorities

Research overview


The broader off-campus recruiting research project

Data collection

  • Method
    • Web-scrape admissions websites
  • Criteria to be included in data collection
    1. Post visits on admissions websites
    2. Organizational type
  • Data collection sample
    • 54 public research universities
    • 49 private research universities
    • 42 selective private liberal arts
  • Data collection period
    • 1/1/2017 to 12/31/2017
    • Ongoing data collection with larger sample

Sample data

Research overview


Focus of this research paper

Research question: what are the similarities and differences in off-campus recruiting patterns across public research universities?

  • Quantitative multiple case study of 15 public research universities


Why this research question:

  • First paper from the larger study
  • Explore behavior inductively, rather than test specific hypotheses
  • Subsequent papers more focused, thematic (e.g., racial red-lining, international recruiting)


Why focus on public research universities:

  • Historic mission of social mobility for state residents
  • Decline in state funding, growth in nonresident enrollment (Jaquette and Curs, 2015)
  • What are they doing to get all these nonresidents? More effort recruiting nonresidents than residents?

Research methods


Defining events

"Off-campus recruiting events" defined as off-campus events hosted by paid staff/consultants focused on soliciting applications


  • Event type
    • Include: college fairs, high school visits, community college visits, counselor events
    • Exclude: admitted or committed student events, interviews
  • Event host
    • Include: paid admissions staff or consultants (e.g. regional recruiters)
    • Exclude: alumni, student volunteers
  • Event location
    • Any off-campus location
    • e.g., high school, community college, hotel, convention center, cafe, etc.

Research methods


Data collection, data processing, data quality

Data collection

  • University website checked four times per year by two staff for URLs with recruiting events
  • Web-scraping scripts run once per week

Data processing

  • "Parsing": transform HTML text into tabular data
  • "Geocoding": use Google Maps API to obtain detailed location data based on limited data
  • Merge recruiting data to secondary data (e.g., schools, communities)

Data quality (are these data any good?)

  • Concern 1: are scraped events properly classified and merged to secondary data?
    • Solution: manually check each scraped event
      • 8 of 15 universities checked thus far
  • Concern 2: are all events posted on admissions website?
    • Solution: issue public records requests for all off-campus recruiting events
      • Received data from 7 of 15 universities; not yet incorporated

Research methods


Secondary data

Secondary data Sources:

  • NCES Common Core of Data (public high schools)
  • NCES Private School University Survey (private high schools)
  • U.S. Census American Community Survey (community characteristics)
  • IPEDS (community colleges)
  • EdFacts Initiative (public high school academic achievement)
  • Equality of Opportunity Project (university income distributions)

Research methods


Research design for analyses

Quantitative multiple case study design

  • "Quantitative case study" uses quantitative data as sole source of evidence (Korzilius, 2010)
  • "Within-case" analyses of recruiting patterns
    • Situate within local context; national overview; "deep dive" of in-state and out-of-state patterns
    • Simple descriptive statistics (e.g., counts), static visualizations, interactive maps
  • "Cross-case" analyses
    • Descriptive statistics and qualitative coding methods


Comparison to alternative research designs

  • Large-N, random sample design
    • Not possible because recruiting data unavailable for random sample
    • Not desirable for present research because large-N designs pool results across cases
  • Qualitative case study (usually collect data from multiple sources)
    • Our design is narrowly focused, systematic analysis of particular phenomena
    • Unlike Stevens (2007), we do not develop a holistic understanding of recruiting behavior

Research methods


Analysis sample

Analysis sample consists of 15 public research universities

  • Chosen from larger data collection sample (N=54) based on "completeness" of recruiting event data
  • Subsequent drafts may reduce sample size based on principles of purposeful sampling (Patton, 2002)
University Sample Characteristics

Deep-Dive Results


Click on a university to take a closer look at the results  


N refers to total number of off-campus recruiting visits

National Overview

 

Select another university

In-State Results

State Map

Select another university

In-State Results

State Figures

Select another university

In-State Results

Metro Area Map

Select another university

In-State Results

Metro Area Figures

Select another university

Out-of-State Results

Top Visited Metro Areas

Select another university

Out-of-State Results

 

Select another university

Out-of-State Results

Metro Area Map

Select another university

Out-of-State Results

Metro Area Figures

Select another university

Out-of-State Results

Metro Area Map

Select another university

Out-of-State Results

Metro Area Figures

Select another university

Cross-university comparisons


Map of Visits

Cross-university comparisons


Number of events by event type

Summary


Summary of results and next steps for this paper

Summary of results

  • Majority of universities in our sample hosted twice as many out-of-state events as in-state events
    • Out-of-state events focus on affluent public schools and private schools
  • Several universities focus more on in-state recruiting (e.g., U. Nebraska, North Carolina State)
  • In-state visits tend to show little evidence of income or racial bias
  • Some universities (e.g., Rutgers) show income/racial bias even in in-state visits
    • Must investigate whether bias persists after controlling for academic achievement, etc.


Next steps

  • Complete data quality checks (e.g., incorporate data from public records requests)
  • Conduct "deep dive" for all universities
  • Compare results across universities
    • Quantitative descriptive analyses; qualitative coding
  • Develop broad categories of recruiting "types" and categorize universities

Future research


Using "data science" and public records requests to study recruiting

Off-campus recruiting project

  • Expand data collection (e.g., regional public universities); publicly release all data
  • Develop manuscripts with more narrow focus (e.g., nexus between state/local politics and visits)

Student list project (collected pilot data)

  • Which student characteristics do universities prioritize when purchasing prospect lists from College Board/ACT?
  • Data collection: Public records requests

Experimental audits of university responses to "inquiries" (pre-pilot stage)

  • More favorable response to inquiries with certain characteristics (e.g., affluence of zip-code/school, evidence of third-party grant/loan)?
  • Data collection: Automate emails; auto-fill "inquiry forms"


Impact goals

  • Change national policy discourse on access inequality
  • Empower local actors to hold universities accountable for access commitments (example HERE)
    • Unless we document enrollment management behavior, we invite symbolic responses

References


 

[1] A. Hanson. “Do college admissions counselors discriminate? Evidence from a correspondence-based field experiment”. In: Economics of Education Review 60 (2017), pp. 86-96. ISSN: 0272-7757. DOI: https://doi.org/10.1016/j.econedurev.2017.08.004. URL: http://www.sciencedirect.com/science/article/pii/S0272775716304526.

[2] M. M. Holland. Divergent pathways to college: Race, class, and inequality in high schools. New Brunswick, NJ: Rutgers University Press, 2019.

[3] O. Jaquette and B. R. Curs. “Creating the out-of-state university: Do public universities increase nonresident freshman enrollment in response to declining state appropriations?” In: Research in Higher Education 56.6 (2015), pp. 535-565. ISSN: 0361-0365.

[4] L. Killgore. “Merit and Competition in Selective College Admissions”. In: Review of Higher Education 32.4 (2009), pp. 469-488. ISSN: 0162-5748; 1090-7009. URL: <Go to ISI>://WOS:000266737500002.

[5] H. Korzilius. “Quantitative Analysis in Case Study”. In: Encyclopedia of case study research. Ed. by A. J. Mills, G. Durepos and E. Wiebe. Thousand Oaks: SAGE Publications, Inc., 2010, pp. 760-764.

[6] M. S. McPherson and M. O. Schapiro. The student aid game. Princeton, NJ: Princeton University Press, 1998.

[7] J. W. Meyer and B. Rowan. “Institutionalized organizations: formal structure as myth and ceremony”. In: The American Journal of Sociology 83.2 (1977), pp. 340-363.

[8] M. Q. Patton. Qualitative research and evaluation methods. Thousand Oaks, Calif.: Sage, 2002. ISBN: 0761919716 9780761919711.

[9] M. L. Stevens. Creating a class: College admissions and the education of elites. Cambridge, MA: Harvard University Press, 2007, p. 308 p. ISBN: 9780674026735 (alk. paper) 067402673X (alk. paper).

[10] The White House. Commitments to action on college opportunity. Tech. rep. The Executive Office of the President, 2014.

[11] The White House. Increasing college opportunity for low-income students. Tech. rep. The Executive Office of the President, 2014.

[12] J. Thompson. Organizations in action. New York: McGraw Hill, 1967.

[13] T. Thornhill. “We Want Black Students, Just Not You: How White Admissions Counselors Screen Black Prospective Students”. In: Sociology of Race and Ethnicity 0.0 (). DOI: 10.1177/2332649218792579. URL: https://journals.sagepub.com/doi/abs/10.1177/2332649218792579.