History of women’s sizing pt.1

Considering the question of the evolution of women’s sizes, it’s very difficult to get a baseline of where we started in order to compare it to today. The first of several reasons is that -as far as I’ve been able to tell- an official survey of women’s bodies was never done until the forties. The only body of work from which to derive information, is that which is gleaned from the sizing charts printed in vintage pattern drafting books. As a point of comparison, below is a chart showing the most typical range of women’s sizes today. The pink box reflects the “average” woman.

While the same could be said of men’s sizes, the historical record is more detailed -of men- because men were more likely to be paying for suit and coat making by tailors; the latter established enterprises that kept formal records, receipts of payments and client measures. Furthermore, the leading tailors of the day were the ones who wrote most of the pattern drafting books. Accordingly, there’s more empirical data available for men’s sizes than women’s. Women were more likely to have clothing made by seamstresses; the latter having been less inclined to have established enterprises as compared to tailors of the day. This is why the Shelton study of the 1940’s -albeit rife with error considering the sampling selection- was so pivotal.

The historical data on women’s sizes that we do have is a poor model of comparison for other reasons. First is styling. We’ve gotten so caught up in the sizing controversies of today that we forget that sizing controversies have raged over the past 100 years. As a point of comparison, below appears a sketch of two women. The woman on the right wore a corset, the other did not. It is highly inappropriate to use the measures of women who wore corsets as a point of comparison, to the measures of women today.

A second reason is demography. As I mentioned above, the documentation of women’s sizes was limited to those women who had the disposable income to have clothing made by (mostly) tailors rather than seamstresses. Still, there is some information on women’s sizes -compiled by women, for women- which I’ll show you tomorrow. Today, I’m showing one sample of the problems with the documentation of women’s sizes as gleaned from a tailor named Morris who wrote a book called Ladies’ Garment Cutting and Making. As you’ll see, his sizing charts reflect the dimensions of women who wore corsets. I should also mention that Morris published the 3rd edition (the one I’m referencing) in the early forties but personally, I have many doubts that he updated his sizing charts in the early forties. There are too many other sources of measurements that contradict his charts. Below is the sizing chart as it appears in his book. You can also find this online.

Below is a comparison of today’s measurements and Morris’

Tomorrow I’ll be comparing Morris with other sources of the same time period. For the time being, note the size breaks that occur in both the smallest and largest of sizes. If you think about it, the sizing controversies of today follow the same pattern. The greatest disagreements and disparities occur in the smallest and the largest of sizes.

Please refer to the other articles in this series which offer substantive supporting material. Add to the discussion rather than backtracking to topics discussed elsewhere. It is likely that the exceptions you’ve thought of have been dissected in depth. For your convenience, links open in a new window or tab.

The Myth of Vanity Sizing
Fit and Sizing Entropy
Push manufacturing; subverting the fit feedback loop
Sizing evolution
Shrinkage and fit
Alternatives in Women’s sizing
Tyranny of tiny sizes?
The history of women’s sizing pt 1
The history of women’s sizing pt 2
The history of women’s sizing pt 3
Sizing is a variety problem
The birth of size 10?
Vanity sizing shoes
Tyranny of tiny sizes pt.2
Vanity sizing: generational edition
Vanity sizing: generational edition pt.2
Vanity sizing: the consumer spending edition

Get New Posts by Email


  1. Julie L. says:


    I suppose a related issue might be the shifting demographic balance over time of how many (and which) women bought ready-made clothing instead of having it custom-fitted or sewing their own. If (as a random and possibly inaccurate guess) ready-made clothing was initially inaccessible to the very poor, then they might’ve been excluded from the sizing stats, frex.

  2. Sherry says:

    The studies don’t seem to account for changes in height (caused by changes in diet, nutrition and lifestyle)?

    Increased economic prosperity, food processing and storage methods, and other economic and technological changes after World War 2, had a material effect on a previously unprecedented proportion of people in this country. Previously, many many people knew what it was like to go hungry. If food wasn’t available, you didn’t eat. If you didn’t eat, you didn’t grow as tall. I would guess that for all of human history, the average heights weren’t as high as the “average” heights of today — which didn’t happen until after the mid-twentieth century.

    Looking at only the bust and waist measurements leaves out a huge amount of very important information. It’s like describing and illustrating a sphere by only discussing circumference and drawing a circle.

    Maybe making customization work is the way to go. The graphs themselves show how narrow a segment is POSSIBLY served by sizing for the “average” woman.

    As a lay person, I am curious — why can’t custom and made-to-order work on a larger scale? I suspect the industry doesn’t need new tools, but new processes. Would any industry people be willing to explain to this lay person? (I’ll ask the question in the FORUM.)

    Something tells me that “everyone” will say it can’t be done, but that someone can and actually will do it successfully and “revolutonize” the industry.

  3. Kathleen says:

    What is the source of this chart?
    It’s empirical, the known range of the variation of human bodies. Even so, it’s not complete. To substantiate the really off-kilter measures, I’d need a data set. For example, I worked with a fit model who had a 28″ bust and a 29″ waist (still doing a burn over the selection of *that* model) but how can I prove that? Unfortunately, much of the information is empirical. There is little hard data. Even among the books I’ll be pulling out, the measures there are based on whichever author’s subjective judgements.

    Are you starting to get an idea of why we don’t have any hard information to go by? Historically, sizing’s always been a matter of opinion and it’s really no different today. Oh, but the pink square comes from the Sizing USA study generated by TC2. I’ve written about those summary results before.

    Looking at only the bust and waist measurements leaves out a huge amount of very important information. It’s like describing and illustrating a sphere by only discussing circumference and drawing a circle.
    Believe me, I am very much aware that we’re leaving out huge chunks of information. For the purposes of simplicity, the graphs need to be clean because I’ll be gunking them up with superimposed plots. What I’m also trying to illustrate is that if we have difficulty managing the range of size possibilities while only considering just two simple dimensions, you can only imagine the range we can muck things up when throwing everything else into the pot. Normally, I prefer to stick to the dimensions of height and weight (and a third factor, the # of survey respondents per “size”) but that example wouldn’t provide the range of confusion or complexity that I want to show with this and upcoming examples. Also, information can be incomplete across sources. None of the pattern books (the empirical historical “data) contain hgt/wgt but they all have bust/waist. Some of the books only provide 3 things, bust, waist and hip so you have to narrow the information or you lose source material entirely.

    Lastly, I’m providing all the source material I have. Feel free to plot whichever other dimensions you’d like.

  4. Eric H says:

    Although the height data would be interesting (and Kathleen almost certainly has those figures as well), of far more import is how much of the population is at each point? Thinking of this like a contour map, if each point had an altitude, would there be a discernible ridgeline? Or several peaks separated by valleys?

    And then … how does it change by age (following a cohort through time)? And how does the entire population change over time?

    A visualization something along the lines of this (load the pdf to really appreciate the data) or this would be nice.

    Ooops, I should be careful what I ask for ;~)

  5. Kathleen says:

    Ooops, I should be careful what I ask for ;~)
    Yep, you better. You’re the one who will be doing up the charts for me (you’d think he’d know better than to make more work for himself).

    We’ve already talked about adding a Z coordinate but I’d like to save that to discuss the weighting of known results we can prove. For example, the sample size (number of respondents) matters. I’ll be showing some numbers from a data set tomorrow of which some measurements should be weighted less and others more, based on the total number of measured participants per clothing size.

  6. J C Sprowls says:

    Hmm… we’re coming back around to my question on gathering sample data. Has anyone that conducted these “studies” reported anything with regard to the costs of the study, method, criteria, or sample size?

    The statistics and analysis portion of the work isn’t difficult; but, the gathering of impartial data could be expensive and challenging.

    Who wants to JAD through this problem offline? I would be interested to see if we, as a community, can define the criteria of several samples to study and publish our results.

  7. Dave K says:

    There are excellent sources of information out there, however most of it is deemed proprietary information and thus not revealed.

    For example you can bet large chains that sell clothes have the sizes of the clothes they have sold in the past so as to predict what to sell in the future. Because their data covers a wide range of age groups and fashions a wide sample is collected. Large chains from Wal-Mart (get away from any personal feeling you have about the company, they are the de facto master of data collection and use) to higher end stores such as Macy’s have this data (sharing it is a separate matter) and I’m sure the US military has a ton of data from the Viet Nam era when the sample was much broader than now – although it’ll be limited to a narrow age group and sex. FOIA anyone?

    But different stores do reflect different economic groups which reduces the sampling error, e.g. the issue of off-the-rack verses seamstresses verses women tailors information collection from prior years discussed in the article.

    This data possessed by stores will yield a very good approximation and inference as to sizes of the population based on what they are buying but not particularly how well they fit or the actual ratios. Also, I’ve lived/worked in the US, Central America, China, and now the Middle East and body types are very very different. so size charts are also dependent on the ethnic mix of the population. Again, I bet a company like Wal-Mart that has stores in virtually every neighborhood in the US and collects data by store their data, if overlaid on a maps from showing income and ethnic mix the result would probably get a very close approximation of size distribution.

    Unfortunately this data isn’t available and still doesn’t resolve the misfit issue within a “size” due to body types ratios.

  8. Marilynn Barber says:

    Speaking of Wal-Mart, I went to Material World in Miami 2 weeks ago and spoke to a dress form company who is working with them because they are trying to figure out why “50% of the customers who come into the stores do not buy clothes”. According to the company some of these retailers are beginning to demand that their contractors buy a company-sanctioned dress form of specific proportions to ensure uniformity in garment sizing. The company is Alvaform and their Size 8 measurements are 36-3/8, 28-1/2, 38-1/2. Size 14 measurements are 40-7/8, 32-3/4, 43.

  9. Carol Kimball says:

    Trivia on why these standards were developed in the ’40s: it was necessary for making uniforms for the men being mobilized to fight (not sure why this didn’t happen in WWI, except a common minor complaint of the guys in the trenches was how badly their uniforms fit). The armed forces had a huge number of bodies available, and in addition to warfare, used the chance to measure them.

    Women’s measurements came as a tag-along for their auxiliary corps.

    The private sector’s tailors and dressmakers benefitted accordinging.

  10. Jane says:

    The standards were developed in the ’40’s because the standardization study was part of FDR’s New Deal to put people to work during the depression.

    Looking at this topic from the ‘half glass full perspective’…..I think it is amazing that retail clothing fits as well as it does. Think about the fact that every single one of us has our own unique shape, size, measurements, lumps, bumps, etc. and, for the most part, we are all able to go into a store and buy clothing to cover our nakedeness….food for thought.

    Another question to ponder. Do you think the death of the dress market is related at all to sizing? Separates are definitely easier to be able to fit.

  11. Sherry says:

    Jane’s comment reminds me of something I’ve been mulling over for a while. I think about how many people around the world, excluding the social elites (i.e. royalty, the warrior class), traditionally dress themselves and how they use fabric. In many parts of the world, the clothes seem to have less seaming, looser cuts (unless the fabric is 1 piece wrapped around the body). There are definitely practical elements involved regarding climate, available materials, and the work a person might do, yet many people still find a way not to look frumpy. Note the case of women’s Hawaiian dress, sarongs, saris, for example, which work well on women of all shapes.

    I think European-style dressing tends to use stiffer fabrics, padding, and much more seaming to create silhouettes. It may be obvious, but I tinker with the idea that perhaps that is why fit becomes so important, because of the WAY clothing is made to fit. Styling almost seems to be in competition — nay, defiance — of the human body. Maybe that is also why sizing (especially for fleshier people) is such a challenge for mass production.

    I think I need to take a trip to a costume museum or find a traditional clothing exhibit somewhere to explore this idea some more. (If anyone knows of any good ones, I’m all ears.)

  12. Julie L. says:

    Cross-culturally, most traditional clothing was also designed to minimize cloth wastage. There’s a fascinating little booklet on the subject called “Cut My Cote” by Dorothy Burnham; it was printed by the Royal Ontario Museum, I suppose to go with an exhibit at some point, but I got my copy through Amazon.

  13. Sherry says:

    Thanks, Julie. I’ll look it up.

    Hmmm . . . less waste, accessible styling . . . maybe there is a way to combine the best of both elements — and still be profitable . . . hmmm . . .

  14. christy fisher says:

    re: “the death of the dress industry” comment:
    It seems separates began to take over around the same time that breast implants were hitting their stride (mid nineties to especially millenium till now)
    I noticed a dramatic body shape shifting with small waist/hip (small sized) women getting implants that were/are not proportiate to their body type.(lots of Cs and D’s on “0’s and 2’s”) Most of these women were/are affluent and style concious shoppers. I honestly think there is comething to the parallel as far as styles.

  15. Kathleen says:

    I think it is amazing that retail clothing fits as well as it does. Think about the fact that every single one of us has our own unique shape, size, measurements, lumps, bumps, etc. and, for the most part, we are all able to go into a store and buy clothing to cover our nakedeness.
    Considering the range of possibilities (which I’ll continue to explain/explore here), I totally agree!

    I love that book, Cut my cote and listed it in my bibliography. Sherry’s comment on european styled clothes is right on the mark too and goes way back to chitons and togas to say nothing of “oriental” dressing. In some off-shoot way, I can’t help but think that the increasing popularity of babyslings is somehow related to all that.

    Another question to ponder. Do you think the death of the dress market is related at all to sizing? Separates are definitely easier to be able to fit.
    Christy says this started in the nineties and I’d meant to respond saying the eighties (the first company I worked for in 1980 did dressy separates) but I didn’t because I suspected that that’s just when I happened to *notice* the advent of separates. I think separates are good, easier; one of the problems with women’s wear *is* having to fit both areas of the body most likely to have disparate measures and men have never had to deal with that issue beyond coveralls.

    Christy, I’ve often wondered about the bigger boob thing but have nothing solid on that. Honestly, I think boob size is incrementally increasing along with height -even among the more slender of women. I think it’s also related to women who are (through education) starting to wear the correct bra size. I have a friend -an accomplished seamstress, got a degree in design and everything- who said she wore a B cup and honestly, I just couldn’t see it, she was bigger than that. She wrote me a couple of months ago saying she’d been refitted and now wears a D which I can totally agree with. There’s been some debate on increasing breast size owing to the increase of hormones in milk -that can’t be discounted either- which can’t be separated from the discussion of the continually decreasing age at menarche.

  16. Alison Cummins says:

    I’ve been looking through a copy of Taschen’s “1000 Nudes, A History of Erotic Photography from 1839-1939” and have been struck by the relative absence of large breasts. (I’m still in the 19th century – haven’t made my way into the 20th century yet.) Most women appear to be a B with some Cs; As are more common than Ds. Is it just that I’m used to the naked, photographed woman being exceptionally well-endowed? Is this a record of a change in the pornographic esthetic or a change in the physique of a population?


  17. jinjer says:

    Is this a record of a change in the pornographic esthetic or a change in the physique of a population?

    Alison, I’ve noticed/wondered the same thing. I wonder if part of it is that, before the advent of bras, larger breasts usually meant lower breasts–perky seems pretty universally erotic. Also, what age used to be considered ideal in women–14? 16? Is it rather new to think of women over 20 as other than Matronly?

  18. Push manufacturing; subverting the fit feedback loop

    “Push” manufacturing can be described as producing an entire line of products without pre-selling and taking orders for it. This means making up a bunch of stuff without knowing if anybody wants it beforehand. In my opinion, push manufacturing is…

  19. trsih says:

    When I ponder fit, I find myself thinking of Charles Frederick Worth. The princess seam is so useful and delightful when it fits properly.

    When I ponder separates, I think of Charles Dana Gibson and how he influenced the acceptance of the shirtwaist.

    I am a fit freak and the current cover of Vogue (with the ladder, poorly posed) is making me crazy. The model on the ladder is wearing such an ill-fitting, poorly cut garment. The armhole if hideous and the fit is … well, not!!! LOL

    Please Vogue, teach the models the history of the ladder in pictures of couture collections, teach the models how to pose properly to imply the history that is (trying to be) told, and please talk to that famous designer and suggest a new patternmaker!!

  20. schrodinger says:

    After reading the entire series of articles on this subject, plus the vanity-sizing series, I’ve come to the conclusion that the Greeks, Romans and to some extent, the Indo-Paks had it right with their ancient clothing: a swath of elegant (or basic) fabrics, tastefully and elegantly draped around the body and then fastened or tucked in place. The style was beautiful and the fit was always right. The only discernible difference between a wealthy person’s clothing and the clothing of a more ‘working class’ individual was the quality and richness of the fabric. A wealthy person would have worn elegantly embroidered garments, perhaps woven with intricate designs and made of finer material, whereas a working person or ‘Average Joe’ probably would have worn a more simple and unembellished garment.

    Ah, to go back to the days of the draped garment! It would certainly be much cooler in Texas (or other hot climate) and no need to buy a new wardrobe in some frustratingly hard-to-find size if one happened to fluctuate by a few pounds! :D

  21. Liz L says:

    Thank you so much for placing this online, it was a fantastic read! Seriously – your site is an amazing resource and a wonderful place to help understand the history of women’s clothing/sizing standards!

  22. Heather Ireland says:

    Can I have your list of sources? I’d like to cite your article and check your facts for a project I’m doing.

  23. ann beck says:

    I happen to be sewing and mending in the 80’s with grandmothers who sewed and mended in the 1920’s. The 20’s clothing was so easy, both of their moms drew the pattern on newspaper, using the largest measurement, basically a rectangle. But closeness of fit makes a huge difference in sizing. In the 80’s loose clothing with belts made altering almost ignorable. But fitting into a 60’s fitted item often involved hoping their was 1/2 seam allowance or you found a bigger size. 1940’s clothing, often had the 1/2 seam allowance and al little more generous cut, because as my grandmother’s said “there was a depression and then a war, most women could not just go buy clothes when they got pregnant or lost weight.” (Yes, there were maternity clothes for later in the pregnancy) So closeness of fit, body shape and shapewear all have to be taken into account. BTW I had one grandmother who was “stout” or a plus size by the time she was 22 and rarely varied more than 15 pounds and one who was average modern size 12 for most of her life. They both would have bought clothes (under garments) from Sears according to their pattern size and made their own clothes, until the 1970’s when it was easier to order clothes than make them.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.