How each score is built
Shul near campus. Straight-line miles from the nearest observant community shul to the nearest edge of the campus where the program is taught, with campus outlines from OpenStreetMap and geocoded shul addresses. Half a mile or less scores 100, one mile 75, two miles 50, five miles 5, six or more 0. Student-only minyanim are shown but not scored. Where no campus outline exists the distance runs from a central building less 0.4 miles, marked est.
Rent. Average monthly rent across the program, two-bedroom until the year you choose, three-bedroom after, divided into your household income. Twenty percent of income scores 100 and sixty percent scores 0.
Size of the observant community. For the 23 cities that also appear in Heshbon's Jewish community affordability index (DASI), we use that index's 0 to 100 rating of shul life. For the rest, we estimate the rating from the number of observant shuls, using the pattern in those 23 cities (score = 38.0 + 10.4 × ln(1 + shuls), n = 23). An eruv adds 15 percent.
Preschool. 328 Jewish early-childhood programs found and geocoded across the 48 cities. The score is 70 percent the best program type within three miles of a community shul (observant 100, Chabad 90, community 80, JCC 75, Conservative 65, Reform 55) and 30 percent how many there are. JCC is present or not. Kosher food counts restaurants, take-out, pizza, cafés and bakeries certified by an observant agency (OU, OK, Star-K, cRc, Kof-K or the city's observant vaad) within about 20 minutes of the main observant neighborhood, recounted in September 2026. Caterers, supermarket counters, ice cream and candy shops, hotels, stadiums and campus dining halls are not counted. The score earns nothing for five places or fewer and rises with the log of the number of kosher eateries up to eighty, plus 15 for a kosher grocery or supermarket department.
Program. Each list uses the best-ranked program in the metro on the current U.S. News list for that field, read from U.S. News in September 2026. A #1 program scores 100, falling evenly to 15 at the bottom of the list. Medical school uses the 2026 research tier (Tier 1 = 100, Tier 2 = 80, Tier 3 = 62, Tier 4 = 45). Schools that no longer submit data are estimated from the parent university's rank, capped at 92. The medical schools we count as the Elite 8 (Harvard, Johns Hopkins, Penn, Stanford, Columbia, Duke, NYU and UCSF) also decline to take part, and we give them full marks. The ten PhD fields use the U.S. News graduate rankings for each field, and the shul distance runs to the campus of the school with the best program in that field. U.S. News shows biomedical engineering ranks only for its top 14 programs, so other biomedical engineering programs are estimated from the school's overall engineering rank and marked est. "Other field" uses the national university rank as a stand-in. Dental schools have no national ranking and all score 70. Health programs measure shul distance to the medical campus. Speech pathology, social work, psychology and the MBA use the main campus. Residency uses the same medical score as a proxy for the academic medical center, since residents are placed by the Match rather than admitted.
What this does not show. Market rents near campus run well above HUD's metro figures in Manhattan, Boston, Palo Alto and Toco Hills. Preschool tuition was found for only a handful of cities. Safety, commuting by transit, and spousal job markets are not scored.