Student Journey Barriers Study

How each persona is built from the data

1. Start from real numbers. Gather what is known about students in the district and the barriers they face, from district figures and national research, with every figure cited in the sources at the bottom of this page.
2. Set the mix. Turn those rates into targets for the set of 50: about 44% food-insecure, 44% carrying a disability, 38% who would place into a developmental course, 16% LGBTQ+, in the district’s real proportions.
3. Build whole students. Distribute the attributes across 50 people so each persona is one coherent student, for example a first-generation, food-insecure nursing hopeful with undiagnosed ADHD, not a trait in isolation.
4. Write the limits, not just the labels. Fill each persona into one fixed template: what she does not know, her device, her reading speed, her language, and the exact moment she would give up, so the agent behaves like her, not like a capable AI.
5. Set it loose on the site. Each persona becomes an AI agent that browses a college’s live public pages from a felt need, thinks aloud, and records where it breaks. How a run works and how people validate it is spelled out below.

Race and ethnicity, district totals only

• White ~43%
• Hispanic or Latino ~37%
• Black ~6%
• Asian ~4%
• Two or more races ~4%
• Native American ~2%
• International and other ~4%

---

Study modality, the 50-persona sample

• Fully online, 13 personas ~26%
• Mixed, 14 personas ~28%
• Fully in person, 23 personas ~46%

---

• ~77%
  Part-time
• ~58%
  Female
• ~26
  Average age
• ~55%
  Students of color
• ~47%
  First-generation
• ~43%
  Food-insecure
• ~20%
  Pell recipients
• ~79%
  Work while enrolled, part-time adults

---

What each persona also carries, counted across the 50

• 22
  Food-insecure
• 22
  Carry a disability or neurodivergence
• 19
  First-generation
• 15
  Non-native English speakers
• 10
  Student-parents and caregivers
• 8
  LGBTQ+
• 3
  Veterans
• 2
  Concurrent-enrollment high-schoolers (ages 15 and 17)
• 2
  International students on an F-1 visa
• 1
  Foster youth
• 1
  DACA / Dreamer
• 1
  Refugee
• 1
  Native American
• 1
  Homeschooled

District totals from the district published data (Fast Facts and the college Student Right-To-Know pages); the study modality split follows a national public two-year baseline pending the district Institutional Research. District totals only, never per-college.

Disability and disclosure across all 50 personas

• Registered with DRS, accommodations active 2
• Diagnosed, not disclosed to the college 4
• Suspected or undiagnosed, never identified 15
• Exploring accommodations, nothing registered 1
• No known disability 28

---

Of the 22 who carry something, by type

• Anxiety, depression, or mood 6
• PTSD, some with TBI 4
• ADHD 4
• Learning disability or dyslexia 3
• Autism spectrum 2
• Low vision 1
• Processing with tech anxiety 1
• Exploring, undetermined 1

These are attributes written into the personas, not measured outcomes. Shares are anchored to national data (about 20% of undergraduates report a disability, with ADHD, autism, learning, and mental-health conditions among them) and tuned to a community-college, Hispanic-Serving, first-generation, low-income population where under-identification runs high. the district does not track a disability a student never discloses, and does not track whether a student was homeschooled, so the undisclosed and homeschooled shares here are modeled from national figures, not district counts.

Age, the 50-persona sample

• Under 18, dual-enrollment 2
• 18 to 24 25
• 25 to 39 17
• 40 and older 6

---

Home language, the 50-persona sample

• English 35
• Spanish 10
• Arabic, Vietnamese, Japanese, Mandarin, Somali 5

Academic preparedness across the 50 personas

• Would place into a developmental (below-100) course 19
• No developmental signal 31

Of the 19, six self-advise or skip placement and risk landing in the wrong course, the placement-and-advising barrier the study tests directly. Share anchored to the national developmental-education rate (about 40%) and tuned to Arizona’s below-average K-12 outcomes; assigned by each student’s story (low literacy, long out of school, first-generation with no one to explain placement), kept distinct from disability and from language alone.

From a candidate to a confirmed finding

1. Run to saturation. Each task repeats at every college with enough personas that new ones stop surfacing new barriers. That is the stopping rule, in place of a fixed quota.
2. Validate with people. A human sample re-runs a share of the tasks and confirms severity (Cohen’s kappa on the agreement). A finding counts only when a person validates it, because severity is the judgment an AI tester gets wrong most often.

• Who the district enrollsDistrict race and ethnicity, age, part-time status, and first-generation share come from the district’s published figures. The food-insecurity share (about 43%, matched in the personas at 44%) is from the Hope Center’s national #RealCollege basic-needs research. the district institutional data · Hope Center basic-needs research
• Disability and under-identificationAbout 20% of undergraduates report a disability, but only around 37% disclose to their college and roughly half are not diagnosed until college, which is why most personas carry a suspected or undisclosed condition rather than a registered one. NCES, students with disabilities · National Disability Center for Student Success
• K-12 preparednessOn the 2024 Nation’s Report Card only about a quarter of fourth and eighth graders were proficient in reading, below the national average, and about half of high-school graduates enroll in college within a year. This is why underpreparedness is written into the set. NAEP 2024 reading, state trends · Arizona Education Progress Meter
• Developmental education and placementNationally about 40% of community-college students take at least one developmental course, and only about 20% referred to developmental math and 37% referred to developmental reading pass the college-level gatekeeper within three years. In the district, below-100 courses are developmental and do not count toward a degree, and placement is largely self-directed through EdReady and the Placement Coach, which is where the wrong-course barrier lives. CCRC, community-college research · New America, who takes remedial courses · the district course selection and placement
• Concurrent enrollment and homeschooled studentsHomeschooled students take college classes as part of high school through concurrent enrollment (a college-taught course on campus or online), not dual enrollment (a course taught at the high school), so they navigate the real campus and site. National data folds them into the concurrent-enrollment count and does not track the homeschool share separately, so their presence here is modeled, not counted. the district concurrent enrollment · NCES, homeschooling
• Gender identity and inclusionAbout 23% of Gen Z adults identify as LGBTQ+ (Gallup 2024), so a young community-college set runs well above the adult average, which is why eight personas are LGBTQ+. For the transgender and nonbinary students the barrier is concrete: a chosen name that does not match the legal name on Canvas, the roster, and email, forms that force a binary choice with no pronoun field, and no clear way to fix either. Gallup, LGBTQ+ identification
• The usability method and AI-agent validationSeverity is rated on Nielsen’s 0 to 4 scale against his usability heuristics, and because AI agents are least reliable at judging severity, trained human raters confirm it and every finding stays a candidate until a person validates it. Nielsen Norman Group, rating severity

The persona attributes are grounded estimates, not measured outcomes. Where the district does not track something, disability a student never discloses, whether a student was homeschooled, the share is modeled from national data and labeled as such.