What the 1930s housing surveys counted in Ohio, 1934 to 1940

Three housing surveys measured Ohio houses between 1934 and 1940, and they are not one series. The federal Real Property Inventory of 1934 covered 64 cities, and the only Ohio places in it were Zanesville and the Cleveland metropolitan district; Columbus was not among them. Columbus took an inventory of its own, enumerated in May 1936 as a work-relief project, issued in 1937 as thirty typed volumes and summarized in a federal report of 1938. The 1940 Census of Housing then counted the Columbus metropolitan district and Franklin County by its own definitions. This page sets out what each counted, in its own words, with the place and the universe on every figure, and never adds one survey to another.

Last updated September 2026 · The Survey of Current Business for November 1934, the Works Progress Administration's 1938 summary of the work-project inventories, two City of Columbus annual reports and the 1940 Census of Housing were read on rendered page images in September 2026 · No prices on this page; the Estimate Builder carries those · Part of the knowledge base

Three surveys, and why they are not one series

Between 1934 and 1940 three different bodies counted Ohio houses, and a figure from one of them does not belong in a sentence with a figure from another. The first was the federal Real Property Inventory of 1934: the Department of Commerce reported that an inventory taken that year with Civil Works Administration funds covered 64 cities, at least one in every state. The second was the City of Columbus's own real property inventory, enumerated in May 1936 as a Works Progress Administration project and issued by the city in 1937. The third was the 1940 Census of Housing, whose Ohio figures the Census Bureau published in Volume II, General Characteristics, Part 4, in a chapter for the state.

Each asked its own questions of its own area. The Department of Commerce's table reports the percentage of a city's dwelling units with a tub and/or shower; the Census Bureau reports counts of dwelling units by whether a bathtub or shower in the structure was theirs alone, shared, or absent. The places differ as well. The 1934 inventory's Ohio figures are for the city of Zanesville and for the Cleveland metropolitan district, and the 1940 census's Central Ohio figures are for the city of Columbus, the Columbus metropolitan district and Franklin County, so no Ohio place appears in both. Columbus's own 1936 inventory is a third thing again, with its own enumeration date, its own area and its own categories. Nothing on this page adds, averages or subtracts across them.

One thing all three agree on, and it is the most useful thing here: none of them counted bathrooms. The 1934 inventory's two plumbing categories were a private indoor water closet and a tub and/or shower. The 1940 census's column head reads bathtub or shower in structure, and its three categories are exclusive use, shared, and none. And the definition of a room these surveys used says that bathrooms, pantries, closets, halls and porches that were unenclosed or very small were not counted among a dwelling unit's rooms, a statement the Works Progress Administration prints in its own appendix and the Department of Commerce prints in a footnote to its own table. The first federal census question asking how many bathrooms a home had came in 1960, and this site's page on bathrooms per house in Ohio carries that series decade by decade.

No statewide Ohio share for 1940 is printed here either; those belong to the page on indoor plumbing in Ohio.

The federal Real Property Inventory of 1934

The Works Progress Administration's Division of Social Research, summarizing these surveys in 1938, called the 1934 federal inventory the first large-scale survey collecting detailed information on the housing situation in the United States, and recorded how it was arranged. The Department of Commerce's Bureau of Foreign and Domestic Commerce put its technique to the Civil Works Administration in the late fall of 1933, on a schedule of 32 questions, with the Bureau of the Census directing the enumeration and the Civil Works Administration furnishing the personnel. The same volume says the first field work was done in Cleveland in December 1933, to test the procedure and the schedule before general field work began in January 1934.

Columbus was not in it. The Department of Commerce's summary table carries one row for each of the 64 cities, and the only Ohio rows are Zanesville and Cleveland; there is no Columbus row at all. The two Ohio rows are not the same kind of place either. The Department of Commerce marked Cleveland with an asterisk, and the note printed above its table says that data for cities marked with an asterisk cover the metropolitan district. The Works Progress Administration's own list of the 64 cities carries the same mark, reading Cleveland with an asterisk and Zanesville without, and that volume says the asterisk marks the 31 cities whose published tables also covered the environs and the whole metropolitan area. That absence, and the project the city ran instead, also appear on this site's page on Columbus building regulation between the wars.

What can be read of the 1934 inventory today is a summary rather than the inventory. The Works Progress Administration says the Department of Commerce published the inventory's tables in a separate volume for each of the 64 cities, so a Cleveland volume and a Zanesville volume exist, and that its own summary prints only the summary totals of the 64 cities surveyed for that reason; the head note to its table sends a reader to the Superintendent of Documents for an individual city's data. Neither Ohio volume was reached for this page. Every 1934 figure below comes instead from the Department of Commerce's article on the inventory in the Survey of Current Business for November 1934, whose table 1 runs across two facing pages and gives each city a single row.

Cleveland had an inventory of its own before the federal one, and the account of it here is one publisher's. The Encyclopedia of Cleveland History says a Real Property Inventory of Metropolitan Cleveland was begun in 1932 by the Cleveland Real Estate Board, which set up a Committee on Real Property Inventory. The same encyclopedia says that inventory cataloged real property by census tracts, that from 1936 its surveys covered all of Cuyahoga County, and that it issued reports from 1933, including a two-page weekly statistical release, with its Bulletin published as late as 1978. The Encyclopedia of Cleveland History also says the Cleveland inventory became the model promoted by the Civil Works Administration in the 1930s and adopted by other cities. That sentence is the encyclopedia's and not the federal record's: the federal volume says only that the first field work was done in Cleveland in December 1933.

What the 1934 inventory counted in Zanesville and Cleveland

The columns decide what a reader may ask of it. On the left-hand page of its table the Department of Commerce recorded, for each city, total persons inventoried, families, structures, dwelling units, owner-occupied units and rental units; on the right-hand page it recorded dwelling units by type, by size in rooms, by the condition and material of the structure, and by the equipment the dwelling units had. Age of structure is on the left-hand page too, as the percentage of structures under 20 years old, which the article says was cut to two age groups, under 20 years and others, because of limitation of space. The equipment group is where the plumbing sits, and it holds two columns only, a private indoor water closet and a tub and/or shower, each printed as a percent of dwelling units. In the article's own words about the 64 cities together, about one-fourth of dwelling units were without a bath tub and/or shower.

In Zanesville, which carries no asterisk and is therefore the city itself, the Department of Commerce inventoried 35,125 persons and counted 10,678 dwelling units, 91.9 percent of them occupied. It counted 10,504 families and 694 extra families there, and put 13.6 percent of the occupied dwelling units in its crowded category.

For the Cleveland metropolitan district the same table shows 1,132,556 persons inventoried and 319,181 dwelling units, 91.5 percent of them occupied, with 313,298 families, 21,107 extra families and the same 13.6 percent of occupied dwelling units in the crowded category. On one item the Department of Commerce made a ranking of its own, writing that Cleveland stood first among these 64 cities for gas for cooking. That is the article's own comparison, it runs among the 64 inventoried cities only, and the Cleveland figure in it is a metropolitan district figure.

Across all 64 cities the Department of Commerce counted 9,074,783 persons inventoried, 2,633,135 dwelling units and 1,931,055 structures, with 2,612,107 families and 183,200 extra families. On condition it wrote that, setting aside the structures needing only minor repairs, 301,783 structures in the 64 cities needed major repairs and a further 43,898 were reported unfit for human use. The table below carries those two classes as percentages, beside the plumbing, heating and lighting columns, for each of the three geographies printed.

Three cautions travel with that table. The first column is a city and the second is a metropolitan district. The third is a total of the inventoried cities and is not a figure for the country. And the four dwelling-unit types in it sum to 88.3 percent as printed; the article does not say what the remaining share was, so nothing is inferred about it.

Columbus took an inventory of its own in 1936

The city's own record of the project is two lines. The City of Columbus's annual report for 1936 lists project 10-25-5052, Real property inventory, among Works Progress Administration projects started but not completed that year, and the annual report for 1937 counts the same inventory among the miscellaneous Works Progress Administration projects completed in that year. Neither report says what it found.

The federal record says a great deal more. The Works Progress Administration's Division of Social Research published a summary of the real property inventories run as work projects in 1938, and that volume carries the Columbus survey, names the City of Columbus as the survey's sponsor, and prints its results. It says the Columbus inventory was enumerated in May 1936 and covered the city together with Bexley, Grandview Heights, Hanford, Marble Cliff and Upper Arlington, which is why its wider column is described here as a metropolitan area rather than as a county or a city. Those suburbs' own water, sewer and bathroom histories are on this site's page on Bexley, Upper Arlington, Grandview Heights and Marble Cliff.

The results were issued rather than printed. The Works Progress Administration records the final report as Real Property Inventory, Columbus, Ohio, 1936, city of Columbus, 1937, a report of text, tables, maps and charts in thirty typed volumes. Most of the other Ohio entries in the same federal list print a page count and the word mimeographed or lithographed; the Columbus entry prints a volume count and the word typed. The same entry says copies went to the Works Progress Administration's research library in Washington, to the Federal Housing Administration's Division of Economics and Statistics and to the City Planning Commission at Columbus City Hall, and that general tables were prepared for the city, for the suburbs and for the metropolitan area including the city.

What the Columbus inventory found

The federal summary prints Columbus in two columns, one for the entire metropolitan area the survey covered and one for the city alone, and the table below is those two columns. Read it as one survey of one date, May 1936, and not as anything that can be set beside 1934 or 1940.

The Columbus survey also split its facilities counts by tenure. In the metropolitan area 2,386 of 32,868 owner-occupied dwelling units, 7.3 percent, had a shared bath or none, against 12,118 of 55,131 tenant-occupied dwelling units, 22.0 percent. On toilets the same two sections show 1,598 owner-occupied dwelling units with no private indoor flush toilet against 5,529 tenant-occupied. Those are the Works Progress Administration's own counts and its own percentages, for the area it surveyed in May 1936.

The inventory also recorded the year each structure was built. The Works Progress Administration's table puts 13,375 of the metropolitan area's 68,454 reporting structures in the years 1905 to 1914, the largest of the ten brackets it printed, and 486 structures in the bracket for 1859 or before. On the condition of structures it puts 22,336 of 68,449 in good condition, 36,996 as needing minor repairs, 7,996 as needing major repairs and 1,121 as unfit for use, and the volume's own definitions say each of those classes rested on the enumerator's judgment.

What the survey counted is bathing equipment available to a dwelling unit, never a bathroom and never a fixture bought or installed. The City of Columbus counted fixtures as well, by type and by year, on the plumbing permits its own office issued; that separate record is set out on this site's page on Columbus plumbing permits and fixtures by year, where a count of water closets or bathtubs is likewise not a count of bathrooms or of houses.

A private bath and a private indoor flush toilet, the two facilities these surveys found dwellings going without, are among the ones the federal property standards behind an insured mortgage set requirements for after the war. What those standards asked of a bathroom, edition by edition, is on this site's page on the FHA and VA property standards and the postwar Central Ohio house.

The 1940 Census of Housing in Columbus and Franklin County

The Census Bureau published Ohio's 1940 housing figures in Volume II of the Census of Housing, General Characteristics, Part 4, in a chapter for the state carrying statistics for Ohio, for large cities of 50,000 or more, for metropolitan districts, counties, urban places and rural areas. Two of its tables cover Central Ohio: one carries the Columbus metropolitan district and the city, and one carries Franklin County and the places inside it. The metropolitan district is not a political unit. The Census Bureau's own introduction says a district was set up for each city of 50,000 or more and takes in the adjacent and contiguous minor civil divisions or incorporated places with 150 or more people per square mile, and that there were 140 of them in 1940 against 96 in 1930. So the Columbus district is neither the city nor the county.

For the Columbus metropolitan district the Census Bureau counted 103,263 dwelling units. Of the 101,317 that reported bathing equipment, it recorded 82,951 with a bathtub or shower in the structure for their exclusive use, 5,877 sharing one and 12,489 with none. Of the 101,314 that reported toilet facilities, it recorded 86,220 with a flush toilet in the structure for their exclusive use, 5,585 sharing one, 194 with a non-flush toilet in the structure, 8,933 with an outside toilet or privy and 382 with no toilet or privy at all. Of the 101,705 that reported water supply, it recorded 96,035 with running water in the dwelling unit, 1,098 with a hand pump in the dwelling unit, 1,833 with running water within fifty feet, 2,290 with another water supply within fifty feet and 449 with no water supply within fifty feet. Each of those three universes is the dwelling units reporting that item, which is fewer than all 103,263.

The county table counts differently. In Franklin County the Census Bureau counted 109,737 dwelling units, and this table prints the units that did not report an item as a category of their own rather than netting them out of a reporting base. On bathing it recorded 84,939 dwelling units with a bathtub or shower for their exclusive use, 5,901 sharing one, 16,712 with none and 2,185 not reporting. On toilets it recorded 88,267 with a flush toilet in the structure for their exclusive use, 5,611 sharing one, 214 with a non-flush toilet in the structure, 12,889 with an outside toilet or privy, 569 with no toilet or privy and 2,187 not reporting. On water supply it recorded 98,514 dwelling units with running water in the dwelling unit, 2,095 with a hand pump in the dwelling unit, 2,179 with running water within fifty feet, 4,401 with another water supply within fifty feet, 776 with no water supply within fifty feet and 1,772 not reporting. Each of those three blocks sums to the county's 109,737.

The census also combined repair with plumbing in one block of the county table. It recorded 84,596 Franklin County dwelling units as not needing major repairs, and of those 70,524 had both a private bath and a private flush toilet, 2,299 had a private flush toilet and no private bath, 6,113 had running water and no private flush toilet, and 5,660 had no running water in the dwelling unit. A further 14,393 dwelling units were recorded as needing major repairs, of which 7,191 had a private bath and a private flush toilet, and 10,748 more did not report repair or plumbing at all. The Census Bureau's own introduction explains that last group: the state-of-repair figures are limited to dwelling units for which every item in the combination was reported, so the 10,748 are units that missed at least one item, not units whose condition was unknown. Its instruction to enumerators defined needing major repairs as parts of the structure such as floors, roof, plaster, walls or foundations requiring repair.

The two tables agree with each other about Columbus. The county table gives the city 86,752 dwelling units, of which 1,738 did not report toilet facilities and 1,720 did not report bathing equipment; take those away and the results are the 85,014 and the 85,032 that the large-city table prints as reporting. The county table also reports five Franklin County places separately, Bexley, Columbus, Grandview Heights, Upper Arlington and Westerville village, with the rest of the county split into rural-nonfarm and rural-farm areas. The city's own 1940 figures at the neighborhood level are carried on this site's page on the King-Lincoln District, and what the city did about substandard housing in the decades that followed is on the page on Columbus housing code enforcement.

What the surveys meant by their own words

The definitions decide what any of these figures can be asked, and all three surveys printed theirs. In the Works Progress Administration's appendix a dwelling unit means a room or group of rooms intended for the occupancy of one family or household as its home and sleeping quarters, and hotels, clubs, rooming houses and institutions were excluded from every one of the surveys.

Bathing equipment, in the same appendix, meant a bathtub or a separate shower, and a bathtub and a shower in the same room were counted as one unit. A private indoor flush toilet meant an indoor toilet with running water, in a bathroom or in a room of its own, and no chemical or outside toilet of any kind was counted. Those are the definitions behind every bath and toilet figure this page carries from the Columbus inventory, and the volume says they are behind the 1934 federal survey's figures as well.

The Census Bureau's words for 1940 are its own. Its definitions volume says a dwelling unit is reported as having a bathtub or shower if either type of equipment, supplied with running water and not necessarily hot water, is available in the structure for the use of the household, and that bathing equipment for the exclusive use of one household is reported separately from equipment shared with one or more other households. Because the question was about equipment anywhere in the structure, a tub shared by several dwelling units was counted once for each of them. The same page defines a room so as to exclude bathrooms, closets, pantries, halls, screened porches and unfinished basement or attic rooms.

Two columns of the 1934 table are easy to misread without the footnotes under it. The Department of Commerce defines an extra family as families sharing dwelling temporarily with usual occupants, and its crowded column combines three degrees: crowded at 1.01 to 2 persons per room, overcrowded at 2.01 to 3 persons per room, and greatly overcrowded above 3. The same footnote carries the Department of Commerce's own statement that bathrooms are not counted as rooms. As for how an extra family was found, the Works Progress Administration prints the enumerator's question: whether anyone living there planned to set up a home of their own when business picks up and jobs are available. The volume says the purpose was to find how many additional dwelling units a city would need when conditions improved.

What the record does not show

The two Ohio volumes of the 1934 inventory were not reached. The Works Progress Administration's summary makes an individual city's data a request to the Superintendent of Documents rather than a download, and searches of the open full-text collections returned neither volume. So the 1934 Ohio figures here are the single summary row each city was given.

The thirty typed volumes of the Columbus inventory are in the same position. The federal summary says where copies were deposited in 1937, and nothing found online holds them. What is on this page is a federal summary of a survey whose own report has not been read.

No dwelling figure for Bexley, Grandview Heights, Upper Arlington or Westerville village is printed here, although the Census Bureau reported all four separately in 1940, because on the scan available for this page two of those columns cannot be read at any magnification. Nothing was found that states what the Columbus Metropolitan Housing Authority's own early surveys counted, either.

Two kinds of figure in these documents are deliberately absent. All of the tables carry rent and property-value columns, and no money figure of any kind appears here. And both the 1936 Columbus inventory and the 1940 census classify households by race in the wording of their own time; no figure from those tables is printed here. How every page in this knowledge base is sourced and checked is set out on the page about where the knowledge base's facts come from.

None of these surveys describes a particular house. They counted dwelling units on one date, by categories about equipment rather than rooms, so a figure from any of them is a statement about a city, a district or a county in a given year, and about nothing smaller.

Sources

Every figure, date and quotation on this page comes from one of these documents, read on the date given. Where two of them disagree, the page says so.

• U.S. Department of Commerce, Bureau of Foreign and Domestic Commerce (Survey of Current Business), The Real Property Inventory of 1934, Survey of Current Business, November 1934, page 16 onward, with table 1 across two facing pages, November 1934; read September 2026.
• U.S. Works Progress Administration, Division of Social Research, Urban Housing: A Summary of Real Property Inventories Conducted as Work Projects, 1934-1936, by Peyton Stapp (Works Progress Administration, Division of Social Research), 1938; read September 2026.
• U.S. Census Bureau, 1940 Census of Housing, Volume II, General Characteristics, Part 4 (Nebraska-Pennsylvania), Chapter 7: Ohio, 1943; read September 2026.
• U.S. Census Bureau, 1940 Census of Housing: Volume 2, General Characteristics (publication index page), 1943; read September 2026.
• Case Western Reserve University (Encyclopedia of Cleveland History), Real Property Inventory of Metropolitan Cleveland (RPI), undated; read September 2026.
• City of Columbus (Columbus Metropolitan Library digital collections), Annual Reports, City of Columbus, Ohio, 1936, the page listing W.P.A. projects started but not completed during 1936, undated (report for 1936); read September 2026.
• City of Columbus (Columbus Metropolitan Library digital collections), Annual Reports, City of Columbus, Ohio, 1937, the page headed ANNUAL REPORT FOR YEAR 1937, W. P. A. WORK, undated (report for 1937); read September 2026.
• U.S. Census Bureau, 1940 Census of Housing, Volume II, General Characteristics, Part 1: United States Summary, Introduction and Definitions of Terms, 1943; read September 2026.

Common questions