A toddler sizing rant from a frustrated mom

This mom's blog about twins Hannah and Sophie gives another perspective on the children's sizing problem. I feel her frustration, and I can understand because as a pattern maker and designer I have difficulty getting the sizing right too. The comments on that particular entry suggest she start her own business selling clothes that fit.

If she were to start a business selling toddler slim clothes, she would discover the same difficulties as other children's fashion designers. Sizing is a big problem that falls all over the anthropometric map. Adult sizing is rather stable in the sense that adults reach a size and tend to stay there - disregarding weight gain or loss. You can categorize adults into sub-groups with similar body proportions and sizing. You can do that with children too but the sub-groups are transitional. Children grow as they age, changing from one group to the next in progression. Ideally, children's clothing allows for that transition in a smooth fashion. In reality, children will occasionally have clothes that are too short, too long, too wide, etc. It is impossible to accommodate them all because they don't all start from the same place. Genetics and individual circumstances play a big part.



This not offered as an excuse for the industry. Surely some creative DE could provide clothing for Hannah and Sophie that is comfortable and fits well. I suspect that the twins were premature because of TTTS. The girls appear to be thriving and doing well. Children born prematurely are a unique sub-group that falls outside the norms. They tend to be smaller and grow differently than other children. It would be an interesting challenge for someone to come up with a product line to help them. Any one up for it?

Comparing pattern shaping and children's sizes

As many children's wear designers know, children's clothing has a lot of sizes. It can become quite the dilemma when trying to decide which sizes to offer. Some DE's offer their styles in as many as 21 sizes. Way back in 2006 I suggested a theory to reduce the number of sizes by combining or eliminating some of them. You may want to go back and review the entry Too Many Sizes and other related entries to see how I have arrived at today.

Anyway, I have tried to put the theory into practice and I have made some progress. I only offer my styles in sizes 3M to 6x. I am still working on the grades for the 4-6x styles, so I am nearly there. My sizes break down like this:

3M, 6M, 9M*, 12M, 18M

24M/2T, 3T, 4T/4

5, 6, 6x

I don't really consider the 9M as a true size. It is a half size between the 6M and 12M and is graded by splitting the grade between the 6M and 12M. The 24M and 2T are essentially the same as are the 4T and 4 - those sizes have been combined for grading purposes. My website still delineates the combined sizes as separate sizes.

This blog entry is not a discussion on the why and wherefores of children's sizing - a surprisingly complex and controversial topic. Instead, I wanted to show a possible grading/pattern problem that shows up now and then. I have been grading and comparing my basic bodice blocks. You should do this too because someone will eventually see the problem and it will be more difficult to fix.

I have drafted my basic bodice block three times, in each of my sample sizes for each of my size ranges - 12M, 3T, 5. (BTW, you can't use the same pattern piece and grade it in all the sizes. Believe me, that is one large headache). The next step is to grade each range separately. Keep in mind that each sample size will have slightly different shaping, but the general shape and proportion should be related.

Before getting too far, the outer fringes of each size range should be compared. For example, the size 18M should be smaller than the 24M/2T and the size 4T/4 should be smaller than the 5. Originally, I had graded the size 4T and 4 separately. In other words the 4T was based off the 3T sample and the 4 was based off the 5. The reason I combined the 4T and 4 was because the shape and overall size was so similar it was a duplication in effort, and I also ran into the problem where the 4T was actually larger than the 4. If you do separate out the sizes than the 4T must be smaller than the 4. If you don't check your grades, someone will bring two dresses to you and say the patterns are wrong, size labels are switched or some other problem.

In the photos below you can see the problem more clearly. In the top picture, the size 4 is laying on top of the 4T. You can see the 4 is smaller than the 4T. In the bottom picture the 4T is on top of the 4 and it is clearly longer with a larger armhole.


To solve this problem, I have been reworking my toddler patterns. I started by combining sizes 4T and 4 so I have one less size to grade. Next my toddler bodices were redrafted to have a shape similar to the 5 (less boxy, smaller armhole). Finally, I regraded the toddler patterns. This is still a work in progress, but the results are much better - each size is progressively larger.

I had the same problem with my size 18M and 24M/2T. In this case it wasn't a grading problem. Instead my infant patterns were proportionally too long compared to the toddler. I fixed this by shortening the bodice slightly.

Anyway, the point is that you should make sure and check the sizes on the outer fringes of each size range and make adjustments so that each size is progressively larger. You can adjust the grade rules (much easier in CAD, btw) or change the shape of the patterns.

(I am ignoring the idea that in the real world an 18M child could be larger than a 24M child. If that is the case, a parent would buy a larger size than 18M and just complain about the craziness of US sizing standards. When I did private label programs for the big box retailers their grade/POM charts progressively got larger with each size. Logically it makes sense even if reality is very different. Anyway, you can allow your sizes to overlap if you want, you'll just need to have an explanation as to why when a sewing contractor becomes confused.).

Grading Complex Styles pt. 2

Part 2 of this series continues the discussion of how to grade complex styles. It is not necessary to have the Handford book to understand the concept, but some experience with grading would be helpful.

This part of the series was going to contain a lot of explanation with lots of examples. There are two reasons that I held off on that. The first is I am considering writing a book on the subject. I am sure it will be a best seller and I will be able to retire early [eye-roll]. The second is that it would be impossible to discuss every possible scenario. It would be best to explain the principle and let you work out your specific grading problem on your own. You will benefit with having to struggle with the material and learn lots in the process.

Just a side note, these pieces have been simplified, no darts or extra gathers. I am also ignoring the seam allowances. The grading is done the same way regardless. This demonstration only shows how to figure out a length grade. The same principle works for widths.

Anyway, let's move on to the examples....

In order to grade a complex style, you need to have basic blocks that are already graded. On top I have a basic block that is already graded. Below that is my ungraded pattern pieces for a bodice with a mid-riff.

There are many ways to determine the total length grade. In this case, I just want to know what the length difference is between the graded basic block and the ungraded pieces. I want to grade my pieces up to the next larger size. To do this, I just lay the ungraded pattern pieces on top of the next size and align the pieces along the horizontal and vertical axis. My drawings are missing the horizontal, but you can pretend there is a line that runs from the side-seam/armscye points across the pieces. BTW if you would rather, you can figure all this out mathematically....

Once you know the length grade difference, you can then decide where you want that length growth to occur on your ungraded pattern pieces. In the examples below, you can have that length split between the two pieces or all of it in one piece or the other (shown in red).

The question you have to ask is:

This is where the beauty of computerized grading happens. I can play with the grade as much as I want - plugging in different numbers until the proportions look the way I want for each size. Hand grading would be a little more complex. The more complex the style, the more difficult the grade can become when doing it by hand. Handford's suggestion to tape the pieces together with long strips of tagboard can work in hand grading. The grader will have to determine if additional grade points should be added and how much movement should occur. This can be done by using Handford's red lines of distribution. Sophisticated manipulation at each grade point can only occur in CAD.

Still the principle works the same regardless of how complex the style or method.

1. Determine your total length/width grades.
2. Calculate the length/width difference between sizes at the grade points.
3. Decide where the growth should occur by comparing the total grade and the difference between each size at the needed grade points. What proportionally looks correct? How is my fit affected?

Grading Pants Notes pt. 3

The following is my final entry on grading infant-toddler pants. Having a copy of the Jack Handford grading manual will be helpful in understanding what these notes mean, especially for this last entry. Hopefully you have at least read the introduction and the instructions for grading a bodice. If you don't have a copy of the book, save these notes anyway - they may come in handy. Part 1 contains an explanation of direction arrows, Part 2 explains notation.

My infant-toddler pants are between knee and mid-calf length. I designed them to peek out just below the skirts of my dresses without distracting from the overall look of the dress. I wanted the trim of the pants barely visible below the skirt hem. So what should be the length grade of my pants?

At the top of the page for the pant grade (i.e., pg 218), Handford states that the grade for various lengths can be determined by studying the grading chart and diagram. The grading chart contains instructions for grading ankle length pants. I don't know why, but it wasn't obvious to me what grade steps to alter. I ended up doing the ankle length grade and decided it was grading to much. Did I mention that I ended up grading my toddler pants three times? Anyway, if you want to grade shorts, or any other length, these are the steps to alter:

Move #5 and #9 (up) is the knee length grade.

Move #6 and #8 is the ankle length grade.

If you want knee length shorts, skip moves #6 & #8. If you want mid-calf length, you could try splitting the difference between the knee length and ankle length grade. Any other lengths will be some variation on the above mentioned moves. If the length is closer to the ankle, follow the ankle grade and vice versa for shorter styles. In other words, these moves are flexible and depend on the style you are grading. It also depends on the personal preference of the individual grader. I graded my pants using the knee-length grade.

One reason I had a little bit of difficulty with determining the length grade has to do with my computerized grading experience. Moves 5 & 6, for example, can be combined in one grading point. The grading of children's patterns can be simplified because the shapes and growth is simpler in some ways. Since you can select an individual grading point and put in relative changes, the moves are interpreted a little bit differently. I am a self-taught grader and it takes me a bit more effort to determine my approach.

Any other question on grading pants? I hope these notes are helpful.

Grading Pants Notes pt. 2

The following are my notes on grading infant-toddler pants. Having a copy of the Jack Handford grading manual will be helpful in understanding what these notes mean. Hopefully you have at least read the introduction and the instructions for grading a bodice. If you don't have a copy of the book, save these notes anyway - they may come in handy. This is part 2 of the series. Part 1 contains an explanation of direction arrows.

This will be a short entry and only partially about grading pants. Mostly it is about Handford's notation and what caused me confusion.

In this drawing I superimposed the direction arrows onto a bodice pattern. There is a similar drawing in Handford's book, page 6 and in Kathleen Fasanella's book, pg 174. All movements start from your point of origin which I have indicated in my drawing. Depending on how you set-up your grading, it will be helpful to draw direction arrows on your patterns.

Now notice the little black triangle under the point of origin. Handford uses this triangle in many drawings. I interpreted those triangles to mean the "point of origin" and that is where I messed up.

Compare the drawings for pants of women (pg 77), men (188-189), and children (219). You'll notice that Handford adds or drops those black triangles almost randomly. Some drawings have them, some don't.

The black triangles don't indicate "point of origin". Instead, I think Handford is borrowing the notation from geometry where it means "right angle". Your direction arrows should be perpendicular - at 90 degrees. If you look at the pants drafts (sorry no more drawings for today), you can see his direction arrows are drawn down the center of the pant legs with two black triangles - meaning the lines should all be perpendicular. In my head, I was seeing "point of origin" or (0,0) on an X-Y coordinate plane. Interpreting things this way caused me to move my pattern pieces incorrectly (my own dumb fault for trying to over analyze things).

Anyway, by nesting my pieces I found my error and realized Handford's direction arrows do not always show the point of origin on all of his drawings. They just show direction.

Clear as mud?

Grading Pants Notes pt. 1

The following are my notes on grading infant-toddler pants. Having a copy of the Jack Handford grading manual will be helpful in understanding what these notes mean. Hopefully you have at least read the introduction and the instructions for grading a bodice. If you don't have a copy of the book, save these notes anyway - they may come in handy.

These are the first of my notes on grading pants using the Jack Handford book. Part one of these notes is an explanation about directions and movements of the pattern pieces. Later I will talk specifically about grading pants, shorts, and what I think is a mistake in the children's chapter. Handford's book is a textbook and like most design text books he starts with grading women's styles. The book progresses from woman's bodices, pants, and skirts, to men's styles, and finally to children. When design/pattern making/grading book authors finally get to the children's info, the information becomes abbreviated and incomplete (if they include it at all). Handford's book is different because he does include infant and toddler information and it is more complete than any other grading manual out there. Still, he falls into the same trap. He assumes the student has worked through all of the previous chapters before arriving at the children's chapter. He assumes you have a strong grading background and understand his method completely. But what about those who have no interest in grading woman's clothing and skip ahead to the chapters with the most relevant information? This is what I did.

Now I do have a strong grading background - computer grading. Hand and computer grading are similar and Handford's method is how it is done in the industry. The direction of the grading is done in a similar fashion. The grade rules are similar. The grade steps are the same. But physically moving a hard, paper pattern around versus selecting points and entering relative coordinates is different. Despite that, I have a general sense of how a pattern should grow or shrink between sizes - and that helped me finally understand Handford's method better.

To start out, you will need to compare the grade instructions for women's pants to children. If you turn to page 79 (grading of women's back pant), you will notice arrows with words around each pattern piece. The arrows look like this drawing. If you flip over to the toddler pant grade (page 225), you will see arrows, but no words. It is at this point that Handford assumes you will know what direction is In, Out, Up, or Down. Now study the drawings of the women's pant and the children's pant. Do you see any other differences? There is one major difference that threw me off - I'll explain later...

The next thing to understand are directions. All patterns will move IN, OUT, UP, or DOWN. Handford shows how this works on page 6 in relation to a bodice. The children's chapter doesn't have these extra helps. So I drew arrows like on the right and below to help remind me of grading direction. I added the words to body center or to body side to remind me of how to move the pattern IN or OUT. As you can see, IN or OUT can change depending on whether you are grading a left or right piece (applies to front or back too). Directions may change again when grading darts. Be sure to study styles with darts to see the changes. Infant-Toddler styles do not have darts, so movements will occur like these drawings. Movements UP move toward the body head and DOWN toward the feet. UP and DOWN movements are always the same.

I hope I haven't thoroughly confused anyone yet. I felt like this brief explanation of the notation and movements was necessary before getting into more detail.

If you have questions, please leave them in comments.

Too Many Sizes pt. 2

In my previous blog entry on this subject I discussed combining sizes to reduce the overall number of sizes produced. As I have been developing patterns for my new line of dresses, Prairie Roses, I have attempted to practice what I preach.

I have been doing all of the pattern making and grading for my new dress line entirely by hand. I do use CAD, but only for my employer who actually owns the system. I try to keep everything separate so there are no ethical questions (My patterns have different shaping and grading anyway). Also, I want to experience what most new designers experience since so few DE's can afford a CAD system. It takes soooo much time to do it all by hand, especially the grading. I am getting faster on the grading, but it is very time consuming. I have been using Jack Handford's book on grading and my graded patterns have turned out really well. The grade rules are beautiful - there is no other way to describe it. They are nice and consistent and all of the pattern measurements fall into an acceptable range.

My patterns fall into the infant-toddler size range and I have chosen to combine the 24M and 2T. I also added a 9 mo size, which is considered a half size and not usually included. I decided to let customers tell me whether they want a 9 mo or not with sales. It isn't that big of a deal to split the grade between the 6 mo and 12 mo. If it doesn't sell, it is easy enough to drop it. Anyway, my sizes break down like this.

3 - 6 - 9 - 12 - 18
24M/2T - 3T - 4T

You will notice I combined the 24M with the toddler group. This means my 24M/2T patterns will be based on grading from the 3T. I decided a 24M child has shaping more closely related to toddlers than infants.

In my infant grouping, the sample/base size turns into the 9M. Since the 9M is something I may not keep around, it doesn't make sense to sample in that size. I prefer to use a 12M as a sample size. If you move the 24M back up to the infant grouping and drop the 9M, then the 12M is the right sample size. Also, it is easier to find a more willing 12M child model than younger.

Now here is a dirty little secret in the childrenswear industry. Many childrenswear companies actually sample in the smallest size, 3M. This is because they end up making A LOT of samples and the samples are made from actual production fabric, not muslin. Sampling in the smallest size saves fabric. The dress form I made is actually a 3M, so that is my sample size too. It causes problems with drafting first patterns and grading. As soon as I can afford it, I will be buying a 12M form and sampling in that size.

Toddler Sweater Update

I started this sweater project over a year ago and it is still not finished. I finally blocked the pieces -- I doubt it was helpful when the sweater is made of synthetic yarn. One thing I noticed during the blocking process is how different the pieces looked from regular toddler patterns. There are several things I noticed that made me wonder...

The front piece has a t-shirt, boxy shape. The shoulder length looks too narrow and the front neck drop and width too big. I need to take measurements to see what toddler size this compares too.









The back piece has no curve on the back neck. It is straight. I imagine a back neck curve is difficult to achieve in a knitted sweater. I wonder if this is typical in knitted garments? I don't knit enough sweaters to know. It could account for why the front neck is so big.








The sleeve shaping seems fine, although they do bow out on the sides.











I have read books on knitwear design. One of these days I will draft my own basic knitwear design and try to knit up a sweater. In any event, all that's left on this sweater is to sew the pieces together and knit the neckband. I still haven't decided on the embellishment. Any ideas?

When do you grade sewing patterns?

Amanda left a comment/question on a previous blog about grading. I thought the answer deserved its own blog entry.

I have what is probably a stupid question. Do you do the grading before or after you add the Seam Allowance to the sloper? Part of my brain says before because you are likely to get a better line, the other says after as it would be easier. I am loving your blog, thank you so very much.

Amanda, not a stupid question at all. The only stupid question is the one not asked. There is a lot of confusion about grading and the difference between a sloper and a block. Let's throw some definitions out and then I'll answer your question.

Sloper - a basic fitting pattern used by costumers and home sewing enthusiasts. The pattern does not have seam allowances, which are added later. People who use slopers do all of their pattern making without seam allowances.

Block - A pattern used by fashion designers and pattern makers which forms the basis of future styles. A block is a perfected pattern that has seam allowances. During the pattern making process the seam allowances remain on the pattern.

Many people believe that using blocks (vs. slopers) is an advanced pattern making technique. It's not really. The pattern making process is no different when using blocks, except for not having to add seam allowances. If anything, it saves time in production settings. It is important for the pattern maker to know what the seam allowances are - which are usually drawn onto the pattern anyway. The exception might be CAD patterns which may or may not have the seam allowances drawn. My CAD patterns do not show seam allowances even though they are there. I do all of my pattern making with the seam allowances already on the pattern. Occasionally, I will turn the seam allowances off to check measurements or to match up seam lines of complex pieces. CAD has greatly simplified the manipulation of seam allowances.

Now to answer the question.... Grading should only occur after the pattern has been perfected and the seam allowances have been added. If you add seam allowances after grading you just add more work to your project and you could introduce errors. This is especially true in a factory setting.

Now in theatrical settings, I know this would never occur. A cutter (aka pattern maker) does their pattern making in the size they need and then adds seam allowances during cutting. If they need two sizes of the same style, they will grade the "slopers". Finally, they trace the pattern onto fabric and add the seam allowances while cutting. This system works in the theater but is not precise enough for apparel production.

Resizing Vintage Patterns versus Grading

The same reader from the previous post also asked:

Also, I found some info on resizing vintage patterns http://www.sensibility.com/pattern/resizepattern.htm thought it might be of interest.

I checked out the link and I do like the website. However, Jenny's pattern resizing tutorial is mis-named. She actually is describing a grading process. It is not a very precise method, but perhaps it would work for a custom or one-time project. I would still encourage my readers to learn a better method, especially for preparing patterns for production.

The idea of re-sizing patterns implies there is an inherent sizing problem which should be solved. If you buy a commercial pattern and you need a longer waist, then you would resize the pattern by slashing and spreading it longer to the required measurement. Jenny does describe such a process at the end of her tutorial.

Jenny goes one step further and shows how to alter an adult pattern into a children's pattern. The method probably will work, but it will require too much fiddling. Why not take a basic block the size you need and draft the same style?

I don't think I am being too hard on Jenny. She does have a nice website. Just realize her perspective is geared more to costumers and others who don't want to be bothered with details.

Simplicity 3295 and vintage pattern measurements

A reader asked a very interesting question

I have started sewing for my 20 month old son and some of his friends and was wondering if you have any experience with vintage patterns (1940's, 50's, 60's). Specifically I'm wondering if you have any thoughts on size variations. I'm sure you are familiar with the dramatic shifts in women's pattern sizes (the 1940's pattern size 12 is very different from the current pattern size 12)...I'm wondering if the same shifts occurred in children's sizes. Any thoughts you would be willing to pass on would be much appreciated!

This question was a little difficult for me to answer because I don't use home sewing patterns very often (not for children, anyway). They require too much work to correct. They have a lot of sloppy patternmaking. It is easier to pull out proven blocks and draft whatever style I want. I will buy a commercial pattern only when I have difficulty figuring out a more difficult design, such as a sleeper and then I only refer to it. I almost always draft my own patterns.

At one time, I thought a collection of vintage children's patterns would be pretty neat to have. I started collecting a while a go, but haven't added much to it. See, I don't actually look at the pattern pieces. Instead, I collect the patterns for design inspiration - in other words, I like the pretty pictures. I can draft patterns with the same style using my own personal blocks. This is one of the few vintage patterns I have from 1950.



I liked this pattern because it has an unusual skirt, pleated fullness in back and a gathered front waist. The bodice has a built-in cap/kimono sleeve. Besides, it is cute!

Now the question of sizing differences between then and now is also a difficult to answer. I haven't really done the research to be able to supply a truly adequate answer. I checked the measurements on the back of this envelope against my collection of measurement charts and the 1950 measurements fall within an acceptable range of contemporary measurements. In fact, the body measurements for this pattern were a little larger than today's measurements. I found this surprising considering the extensive reporting about childhood obesity. Of course, this is one style, in one size, from one company. I don't know how other vintage patterns may measure up. The best thing, of course, is to measure your own child and compare the measurements with the pattern envelope charts.

One difference that may exist between vintage and contemporary patterns is the amount of ease, design and wearing. Styles from the 1960's may fit closer to the body versus styles from the 1980's. This may cause one to think the patterns from the 60's were made to fit smaller people versus larger people in the 80's. The reality is that the prevailing styles influenced fit. It is helpful to try and look at the patterns from the correct perspective. Unless you are striving for authenticity, it is worth the time to draft your own patterns.

I haven't really seen a "vanity" sizing shift in children's clothing that is comparable to women's clothing. US Children's sizes are based on age designations. If anything, the sizing has become more specific. The 1950's may have had one infant size and then start with toddler sizing as 1, 2, 3, etc. Now, infant sizing is broken down by age too -- 0-3M, for example. Measurements are derived from the average measurements of children in each age grouping. Some consumers become frustrated by this sizing system because their children may vary from the norm (not to mention every manufacturer sizes differently). They may have a 2 year old child that actually wears a 4T. Trying on multiple sizes of one style can cause frustration in a changing room, if parents even take the time to try clothes on.

European sizing is evolving to a more intelligent sizing system. The Europeans are basing their sizing on height and weight measurements, available on hang tags and labels. Body size measurements are broken down by age grouping too, but they are not given age designations. This concept enables the consumer to buy the most appropriate size regardless of the age of the child.

A Word on Handford and Children's Sizes

I have previously blogged about Jack Handford's book Professional Pattern Grading (I like this grading book!). I just graded my basic infant bodices using Handford's grading charts and I just wanted to pass on a few insights.

If you read my blog on children's sizing (Too Many Sizes!), then you will know that there is a lot of variety in how manufacturer's lay out their size ranges. Handford is not really any different. His sizing breaks down like this:

3M, 6M, 12M, 18M, 24M, 36M

You will notice that the 9M size is missing. A 9M is considered a half size between 6M and 12M. Some manufacturers include it and others skip it. After studying dozens of measurement charts, I tend to favor skipping it. Children grow really fast in the first few months of life, so there is no need to include lots of possible sizes.

Another thing to consider is his 24m and 36M sizes. His toddler sizes break down like this:

1, 2, 3, 4

Except for boutique/specialty shop stores, you don't really see toddler sizes denoted like this. Normally, the toddler range is 2T, 3T, and 4T (some throw in a 5T). Anyway, there may be some overlap between the infant and toddler sizes with your patterns. This doesn't mean you can grade all of the infant and toddler sizes together. You will still need an infant block and a toddler block. You may have some shaping differences between the two.

It took me some time to wrap my brain around his grading instructions. I could look at the diagrams and see that this was the type of grading done in the industry. Even computer grading takes it's cues from this method. The actual grading process, however, varies from computerized grading.

I am accustomed to working with all of the pieces nested together. I can select a grading point and enter in the X,Y changes. It really is pretty simple. Handford has you grade one size at a time. After grading one size, you cut it out and use it to grade the next size. Precision is absolutely key because mistakes gradually increase with each size. I eventually got the hang of it and found it to be pretty easy. Once I had all of the sizes, I nested them on top of each other and I could see how well it worked. And it did work well.

So now I have two sets of infant bodice patterns using the Handford method and an adaptation of Aldrich's measurement charts. I am going to compare the two to see which I like better. So far, there doesn't seem to be too much disagreement between them. I am also trying to decide if I should add in a 9M. Right now, I am leaning toward Handford.

Sizing Studies for Children

A few weeks ago, a new mother complained about the variation in infant clothing sizes. She explained that her 12 month daughter wore clothes ranging in size from 6m to 18m. Some marked for 12 months were either too big or too small. Finding clothes that fit was a matter of trying things on a fussy baby.

She stated, "You can't trust what the size tag says."

Kathleen at Fashion-Incubator wrote an article about problems with sizing studies, especially for women. Children's sizing studies present some of the same difficulties. Children's sizing is complicated by the fact that children grow and at different rates. Ethnicity plays a part in proportions, weight, and growth rates. As a technical designer, I feel that I am only taking my best guess when drafting patterns. I have to synthesize about a dozen different size charts and methods to come up with something that works and fits! I firmly believe that even infant children deserve clothing that fits.

My concern with most measurement charts and sizing studies are that they are minimally useful for patternmaking for infants. Many measurement size charts either completely ignore or present limited info for infant sizing. The Armstrong book starts her charts at size 2/2T (some of her measurements are a bit funky too). The Gloria Mortimer-Dunn book is the same way. Childrenswear Design, has some basic information for infant patternmaking although the information is not complete. The best presentation of measurement info is from Winifred Aldrich in her book Metric Pattern Cutting for Children's Wear and Babywear: From Birth to 14 Years .

I have drafted basic blocks using Aldrich's charts and instructions. They turned out ok, but definitely needed refinement for style and fit. I thought the basic, fitted blocks need a little more ease and a lower neck. The flat blocks needed a better armhole shaping. Despite that, with some tweaking, you could have a nice set of infant blocks to work with. Part of my difficulty could be that the British are accustomed to a closer fitting block and the metric system. I had to wrap my head around metric conversions, which probably introduced some inaccuracies.

I haven't talked much about measurement charts available from ASTM. This is because I have not purchased a set of charts from them. I don't have the need at this point. Other government studies are interesting, but don't contain enough of the measurements needed to draft basic blocks. Growth charts and retail size charts (and a few in patternmaking books), list infant sizes by pounds and lengths. Those are interesting for comparison and not useful for drafting patterns.

I am not sure what new childrenswear designers are to do. Most companies develop their own measurement charts and some are better than others. Getting those first, good bodice block patterns are critical because every future design is based off of them. It may be easier to drape your first blocks off of dress forms rather than drafting from measurements. Or maybe a combination of both. This sizing study problem could explain why infant sizing is all over the place.

Over the years, I have settled with basic patterns that have proven to work. They continue to be modified for improvement and someday they will be what I hope is a great pattern.

Points of Measure and Grading

In a previous grading blog, I showed the first step in developing grade rules. I usually break my measurement and grading charts into two charts. Here is an example:



I usually stop at the middle chart and work the rest of the math out in my head as I grade. Some of you may want to break your measurements down even further. If a pattern piece is symmetrical, you can grade half of the pattern at a time. The third chart shows the grades for half of the pattern width.

My charts have a column on the left titled POM, or point of measure. Those points of measure should correspond to a drawing like this:



The measurements in the top chart above are based off pattern measurements (which are made up for this example), rather than body measurement charts. The grades are developed from body measurements. To grade successfully, you will need to take a lot of measurements off your base size patterns. For a bodice pattern, you will need armhole circumference, across the shoulder neck circumference, neck width, neck depth, and possibly more. Work between your body measurements, pattern measurements, grade, and ease requirements to develop appropriate grade rules for each point of measure. BTW, if your pattern measurements are finished measurements (minus seam allowances), you can use these charts to measure finished garments while doing quality assessments.

There are standard reference grading charts available from various sources. You can refer to them to develop your grade rules, if you like, and it may save you some time. Standard grading charts are usually available for adult, most often, women's clothing. Children's grading charts are available, but I have found them less useful. There are too many variations in children's body measurements, and every company has their own set. I usually have to develop grading charts that are unique for each company. When a designer asks me to grade a pattern, I have to work off of their measurement charts and basic pattern pieces for the grade to make sense. I will compare their measurement charts with my own to make sure their measurement and grades fall within an acceptable range.

If you do use a standard reference grading chart, you will need to double check your pattern/body measurements across the range. Standard grade charts do not accommodate variations in body type that your company may try to fit.

In the next grading blog, I will begin to explain the actual process of grading.

Comparing Butterick 6030 neckline measurement

I didn't realize I had so many boys' shirts in my stash! Each shirt has a slightly different construction technique. They range from casual to dressy. Some have a convertible collar with a facing and others have a traditional 2-piece collar. This is a great resource for me to figure out more of what is wrong with the Butterick pattern.

The first thing I did was take some key measurements off of the size 24M shirts. The button bands averaged 1.125" and I adjusted my new pattern accordingly. The second measurement was the necklines. Both 24M and 2T shirts measured 12 inches, which I think is right on target. Compare these numbers:

Actual neck measurement = 10.125"
Butterick neck measurement = 15" (4.875" of ease!)
My pattern measurement (1st attempt) = 14" (3.875" of ease)
Both RTW shirts neck = 12" (2" of ease, right on!)

To draft the neckline of the pattern small enough, I need to have a basic 24M block made up. I am not going to keep guessing -- it just wastes too much time. This is enough motivation for me to work on my basic blocks again (My New Year's Resolution!). I pulled them out yesterday and made some important decisions. I can officially announce I have a basic bodice block that I like. The next step is to figure out my grade rules and grade my basic block up to a 24M.

Figuring grade rules and grading my blocks will take me some time, so the shirt project will be put off for a little while. In the mean time, I may get back to my promised grading blogs. Hopefully, you won't get too bored by it. Grading is as dry and boring as a topic can get.

Creating a grading standard pt. 3

2019 notes - The original images for this blog entry have disappeared and I haven't had time to recreate them. I do have more complete examples in my ebook, The Organized Fashion Designer.  
 
I finally made my example measurement/grading charts. It took a little dancing as I had to create them in a spreadsheet, export them as a PDF and then convert them over to a jpg. Before I get too far, it may be helpful to review my previous blogs on grading: Creating a grading standard and Creating a grading standard part 2. You should have your own measurement charts handy. BTW, the measurements in my charts are real. I found them from the Sears website. I do not necessarily endorse Sears measurements as the industry standard - they were just handy (To be fair, their measurements are pretty good). Your own measurement charts should be far more complete and detailed than these for pattern development and actual grading anyway. This is the first step I take in developing actual grade rules.

Right off, I hope you notice a few important things. First, I have clearly marked my sample sizes. In this example they are sizes 10 and 10+. Each chart is labeled clearly. You would think these things should be obvious, but you would be surprised at what I have seen. Hanging off of the left side are numbers. These numbers are points of measure and should correspond to a How to Measure diagram (a future blog) and are not relevant for the immediate discussion. Also notice that my size ranges differ. The regular sizes run 7-16 and the plus sizes run 8+ to 18+. This is a fairly typical difference between the two groups. Also notice the difference in the measurements. There should be some obvious differences between a Misses and a Plus sizes chart too.

Usually, I have my grading chart separate from my measurement chart. I combined them here so you could more easily see how I am developing my grading rules. In the column for my sample sizes, I have placed a zero. A sample size is also called a base size in grading. It is your starting point and each subsequent size will grow or shrink proportionally off your base size. Next I subtract the difference between the base size and the next largest size. In the regular size chart, you will notice there is a 1.5" difference in the chest measurement from a size 10 to a size 12. This difference is called a grade step. Next I subtract the difference between a size 14 and a size 12. I don't subtract Size 14 from a size 10 because that is not the next grade step. Continue to subtract the next larger size with the previous size.

To calculate the grade for your smaller sizes, subtract your sample size from the next smaller size. Be sure to add the negative sign, which indicates the grade is getting smaller. Repeat by subtracting each size with the next smaller size. Create a grade for each measurement on your chart.

You will notice that the measurements and grades all have beautiful numbers. The measurements increase or decrease proportionally causing the grades between sizes to be relatively the same. This is where the art of grading and measurements come into play. I can guarantee that actual body measurements are not this pretty or consistent. These numbers have been averaged and rounded and are based on a large body measurement sampling. The numbers have been intentionally made easy to work with. Your measurements should be easy to work with too and you can adjust them as necessary.

What about accuracy? Rounding does introduce inaccuracies in your measurement charts, but only a little. If you look at growth charts, you will find that certain measurements will fit 50%, 80%, 95% or 97% of girls. If you adjust your numbers up or down, you will want to make sure those numbers fall into the 95% percentile. Adult measurements and sizing are similarly developed. Because these measurements are based off of measurement studies, it means a real girl will pick the size that most closely matches her measurements. Your measurement numbers just need to fall within the highest percentage category. You can round to the nearest 0.5" or 0.125", or whatever. Adjust your measurements so that you get a relatively consistent and even grade across the sizes. In my regular sizes I have a consistent 1.5" chest grade and a 2" grade for my plus sizes. Sure, you could throw in what ever grade steps you choose as long as you have justification for it.

Finally, I hope you can see the difference between the regular and plus sizes. Not only are plus size measurements larger, they are proportionally larger (the grade step is larger). This is why you absolutely cannot grade a plus size pattern from a regular sized pattern piece. Keep both categories, pattern pieces, measurements and grades separate from each other. I promise it will save your sanity.

I know this is a lot of explanation for this first step. If anyone has a question about this, just leave them in comments and I will try to answer them. In my next article in this series, I will explain how to create grade rules based off your first grading chart.

Creating a grading standard pt. 2

In a previous blog, I wrote about the first steps in creating a grading standard, but failed to discuss the details. It would be helpful to also review my blog entry titled Too Many Sizes! In this blog, I will try to explain grading terms and the rules by which grading can occur. In the future, I will show how to develop your actual grade rules and how I grade a pattern (which is different from any other method I have seen).

There seems to be a lot of confusion about grading. Some people view grading as a magical process that can turn your beautiful pattern in one size into any other size you desire. Before anyone can grade your pattern, you should have already done some homework. You should have chosen a category, size range, and measurement chart. Your category and size standard should be fairly simple. You should only be working in one category and your size range should not have more than 6 sizes. You may want to work in men and womens plus and regular sizes all at once, but it is just not possible.

Now, I have to stop here for a moment and explain the concept of categories. I don't know if this is the appropriate term or not. I work in a children's category, but you may work in Misses, Juniors, Mens, or whatever. Within those broad categories, there are sub-categories or classifications based on your sizing system. A classification is based on a specific figure type such as Misses, Petites, Plus, or Talls. Each of these designations have a separate and distinct sizing system. You can't magically start with a Misses size and turn it into a Plus size by grading it. That is not how grading works, so don't even attempt it. There is no magic formula.

Grading is a simple, yet difficult concept for people to grasp. Grading is a process by which a base pattern is proportionally changed to create smaller or larger sizes. Grading is always based on a set of measurements specific to a size range and classification. Successful grading does not change the overall proportions of the intended design as the size changes. This is why a size range should be limited to about 6 sizes. Any more than that and the largest or smallest size will be proportionally wrong.

Successful grading is all about having a good starting point. A base pattern is a pattern created and perfected to fit a sample size. From that base pattern, you will create the other sizes in your size range. Now some of you want to offer Plus sizes in addition to your Misses Sizes. It should be as simple as grading your base Misses patterns up to the appropriate plus size, right? After all, it would save a great deal of work. You already have a perfect pattern and why would you want to go to the work of creating a whole new set of patterns.

The reality is, you will have to create a whole new set of patterns in a new base size contained in a new size range and classification. I know this sounds like a lot of work. The truth is that it is, at least initially. You will have to develop a set of base patterns for your Misses sizes and a separate set for your Plus sizes. You may sample a new style in a Misses size initially. Once a style is approved, your patternmaker will then create patterns in any other size ranges you want to offer. The process required to create the plus size patterns will be very similar, if not identical, to the process required to draft the Misses size patterns. There may be a little tweaking for fit during sampling, but the process will be faster than the original style development. Your patternmaker will already know what should be done.

For example, one company I work for will create a new style in a size 5 (in a children's size range of 4-6x). They then decide they want to offer that style in the entire range 2T-16 and plus sizes. I have already broken up the sizes into ranges of 2T-4T, 4-6x, 7-16, and 8+ to 20+. I then make patterns in the sample sizes of 3T, 5 (which is already done), 10, and 10+. The patterns for each sample size are then graded within their range. It may sound like a lot of work, especially if done entirely by hand. Since I work in a CAD environment, I can accomplish this task in less than a day. By hand, it may take 2-3 days.

Going back to our previous example......Once your set of Misses patterns and Plus size patterns are finished, you will then send those off to be graded. And this is where you have to create your grade rules. To create grade rules, you will need your measurement charts. So here is your homework assignment. Read your measurement chart. If you are offering more than one classification, compare the two. How does a Plus size differ from a Misses? Do your measurements make sense? Do the measurements decrease for the smaller sizes and increase for the bigger? In the next blog, I will show how to create your grade rules from your measurement charts. Once you see the grade rules, you will then understand why you can't grade a Misses size into a Plus size.

How to create a care, content and sizing tag for clothing

Roland Russwurm [CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0)]*

Deciding what information to include on a hang tag or a garment's care and content tag can be frustrating. There are certain legal requirements for care and content tags of items sold in the United States and that can be found at the Federal Trade Commission website.

Let's start with the care/content tags. Let's assume that you have complied with all of the legal requirements. What else should go on your care and content tag? The answer is as little as possible. The tag has a finite amount of space, and it must be legible. But there is some critical information that will help the consumer pick just the right size.

Sizing information is one of those things that should be on the tag. For children's clothing it helps to put not only the size but also the weight. It would be nice to also include the height, but I rarely see this done in the US. I have seen the sizing information arranged different ways. For example, it could read like this: 0-3M (7-12lb).

This additional bit of information is very important to consumers. Even though we use a size system that designates an age, children come in all shapes in sizes at each stage of life. Ever hear of a newborn weighing 9lb? A newborn weighing 9lb at birth will weigh 12-15lb at about 3mo, and quickly move into 6mo clothing. Including the weight on a tag helps take out the guess work for the consumer. And don't forget every children's company breaks their sizing down differently. A 3mo outfit from one company will not be the same size as a 3mo outfit from another.

If there isn't room on a care/content tag for additional information, then it is helpful to include it on a hangtag. BabyGap has sizing information available on a card that is the size of a business card and it can be stuck in a wallet. What a great service for the customer! Not to mention, it probably helps reduce returns.

Most moms know the height and weight of their baby. They are given this information when they visit their pediatrician. The doctor will compare the baby's height and weight with a growth chart to make sure the baby is growing properly

If you have an online store, a size chart that includes basic height and weight information will greatly help your customers.

*Photo Credit

Creating a grading standard for children's clothing

Before I can move on to the grading question, I must make a couple of comments on my previous post. The best time to determine a sizing standard is at the very begining of your company. If you have been in business for 30 years, for example, it will be very difficult to change things around. So please, please do your homework.

Another thing, you can make your sizing standard anyway you want. Even though I am suggesting a simplified sizing standard, it may not be appropriate for you. If your intended customer is a big box retailer, it may be too simple. In fact, they may not like it. They are accustomed to choosing from a lot of different sizes. If you become part of a private label program, they will tell you what the sizing standard should be anyway. If you have something similar already set-up, there will be less stress for your patternmaker. Again, the sizing standard suggested in the previous post would be more appropriate for a company targeting specialty boutiques and small retailers.

Ok. Now the grading questiong...

"Can't we use the 3mo size for our base size and grade everything up from there?"

The motivation for this question is to save some work. In essence you make your pattern only one time and let the grading take care of all of the other sizes. While the idea is good, it presents some problems.

First, it is important to know how a child grows. A very good description and diagrams are shown in the book Patternmaking for Fashion Design by Armstrong. In my second edition of the book, it is on page 674. Up to about age 3 children are cylindrical in shape. There is little differentiation between chest, waist, and hips. At about age 4-5, girls start to develop a waist. At age 8, there is more definition and curves. Boys and girls are similar in shape up to about age 8.

Because there is a change in body shape at regular intervals, a patternmaker will break-up the sizes like this:

0-3M to 24M, 2T-4T, 4-6x, 7-16

By breaking up the patterns into these size ranges, a patternmaker would then make base size patterns in a 12M, 3T, 5, and an 8 (or whatever sizes are chosen). You could sample your initial style in any size you prefer - hopefully a middle size. When the design is approved and read to go into pre-production, the patternmaker will then make patterns for your style in each of the other base sizes. Those patterns are then graded for each range.

Some manufacturers will go to the trouble of sampling their style in each range. Retailers like to look at more than one size range because they have to "see" how to merchandise their floor areas. It is possible you will end up making a lot of sales rep's samples if you plan on selling the whole range. These extra samples are called sisters or brothers.

Even with our simplified sizing standard, you would still break up your sizes like this:

0-3M to 3, 4-6, and 7-16

Now some design entrepeneur's may still insist on having only one base size and grading everything off of that. What will happen is you will introduce errors into your patterns in the extreme ends of the range. The grading will be more complicated. The fit will be off in the transitional sizes between infant to toddler, young child to older child. For organization sake, your patternmaker/grader will greatly appreciate having things broken down into simpler groups.

Also, you will be bucking industry standards and not playing by the rules. Walk into any department store and look at how the children's clothes are arranged. You will notice that the clothes are arranged together in the size ranges we talked about above - infant, toddler, young child (4-6x), and older child (7-16). Your order form should have your styles broken down similarly.

By this point you are probably thinking that it is still a lot of work to make that many patterns for each base size. If you have a CAD system with nifty pattern drafting and grading capabilities, it isn't really that big of a deal. Even if you had to do it by hand, it should only take a day or so. Your patternmaker will be just repeating the same drafting process over and over again. If you make dozens of styles every season, you may seriously consider investing in a CAD system.

Too many sizes!

I am currently in the process of rewriting this article for another project. A lot of what I wrote in 2006 is a reflection of my thinking back then. I have since refined my thoughts which will be published later. So please consider this article a bit out-of-date.

One designer asked me two fundamental questions. At the time I wasn't sure how to answer them. The first is really a grading question and the second a sizing standard. Both are related. To keep things simple, I will start with the sizing standard question.

"Why are there so many sizes for children's clothing?"

Because children grow. Ok. That is the easy way out. The truth is that children grow and manufacturers try to have a range of sizes to choose from.

The other half of the question was, "Can we have fewer sizes?"

This is a much more complicated answer. Because children's clothing can cover such a wide range, it is possible to have a lot of sizes. One company I worked for produced clothing from Preemie to a size 16 - not including the plus sizes. When you break this down it looks like this:

Preemie, NB, 3M, 6M, 9M, 12M, 18M, 24M, 2T, 3T, 4T, 4, 5, 6, 6x, 7, 8, 10, 12, 14, 16.

This is a whopping 21 sizes! This complicates bookkeeping and sales information incredibly. If you create a style in one colorway, then each size will have its own SKU (stock keeping unit). If you add a colorway, then each size in this style can have two SKU's - one for each color. The company wanted to try and reduce the number of SKU's each season. This particular company produced a new line every 4 months which consisted of dozens of new styles. Thinking about it can give anyone a headache.

To reduce the number of sizes, you have to either combine some of the sizes into ranges and eliminate some duplication. The duplication is the easiest to see first. There is very little difference between a 24M and a 2T. A 24 month old child is essentially a 2 year old. The T simply means that ease is added to the pant area for a diaper.* Most 24 month old children and 2T children are still wearing diapers, so there should little to no difference in the patterns there.

*I'm not sure where the idea that the T in the size 2T implied the size included diaper ease. I repeated this idea because it was all I knew at the time. After a lot of research, I now believe this idea is a myth that has been perpetuated around the internet. I have found no size study information that supports this idea. Rather it has become a standard custom to refer to toddler sizing in this way. As a pattern maker I always included ease for a diaper in the toddler sizes because they wear diapers! The ease is not much more than what is needed for wearing pants anyway. Diapers do not add much bulk these days. Perhaps in years gone past in the age of cloth diapering, it was more of a concern.

The other likely duplication is the 4T and 4. Again these sizes overlap with only a slight difference in the pant area for a diaper. Most four year olds are potty-trained, although it is possible that some are not. In any event, if you manufacturer girl's dresses, this is another area where sizes could be combined. Some companies have started putting out a 5T, but it would be unusual to find a five year old wearing a diaper.

The next size duplication looks like the 6 and 6x. In this case, there is a fitting difference introduced. There should be little girth difference in the patterns, but there will be a length difference. A 6x is a taller size for a 6 year old. Some retailers combine the 6x with a 7. This is where it would be important to know your customers before eliminating or combining sizes.*

 *(Please note that this paragraph has been modified. I made a big assumption that was just wrong stating that the size 6x is a plus size version of the size 6. This isn't true. It is a taller version of the size 6. In any event, the paragraph above has been edited with the correct information).

So now our sizing looks like this:

Preemie, NB, 3M, 6M, 9M, 12M, 18M, 24M/2T, 3T, 4T/4, 5, 6, 7, 8, 10, 12, 14, 16.

And this is where I have to interject. Retailers, and especially big box retailers, do differentiate the sizing of the patterns on the sizes we just combined. A 2T is just a little bit bigger than a 24M, even though it really shouldn't be. Consumers have become accustomed to this type of sizing system and so have the technical designers in the business. In other words, don't expect any big changes anytime soon. This proposal is a way to simplify things. You are more likely to find a simplified sizing standard in boutique or specialty stores.

Ok, back on track now. The next step is to create a range of sizes. This is simplest in the infant sizes and another grey area. Every major manufacturer has come up with their own size range break-down, and it really is all over the place.
I really like how JcPenney has broken down their infant size range:

Newborn, 0-3M, 6-9M, 12M, 18M, 24M/2T

The nice thing about this sizing standard is that there are no overlapping of sizes.
Some companies will create a range like this:

Newborn, 0-3M, 3-6M, 6-9M, 9-12M, 12-18M, 18-24M

As you can see, there is a lot of overlapping. This could cause confusion for a customer because it is difficult to pick just the right size range. Also, this doesn't reduce the number of sizes carried.

If we eliminate the newborn and preemie sizes and use the size range from JcPenney, we would get this:

0-3M, 6-9M, 12M, 18M, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16.

If you look carefully, you will see that I eliminated the 24M size and the T's on the 2, 3, and 4. This is where you have to decide how the size will appear on the care/content tag or hang tag. It would be best not to confuse customers - keep things simple and logical here.

You will also notice that there seems to be a size missing in the infant sizes. The size 6-9M is often combined because either the 6M or the 9M is considered a half size between the 3M and the 12M. This is obvious when you study measurement charts, so just take my word on it for now.

The other thing to consider when generating a size standard is the patternmaker/grader. The size 0-3M, for example can be created in one of two ways. The measurements of the Newborn and 3M could be averaged out or the patternmaker could just make the patterns a true 3M. In order to fit the most children, it is best to make the patterns fit the high-end of the range. In our example, the patterns would be made in a 3M. The clothing will fit loose on the small end of the range, but kids grow fast. The only exception might be sleepwear which needs to fit snug to the body in all of the size ranges. This is another place where you need to know your product and customer well.

By following this example, we have gone from a whopping 21 sizes to 15! While this still creates a lot of sizes, it is so much simpler and easier to understand.

0-3M, 6-9M, 12M, 18M, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16.

The next question to address is the grading question. And that will have to wait for tomorrow.