Content uploaded by Hwa Kyung Song
Author content
All content in this area was uploaded by Hwa Kyung Song on May 18, 2018
Content may be subject to copyright.
International Journal of Clothing Science and Technology
Women’s petite and regular body measurements compared to current retail sizing
conventions
Youngsook Kim Hwa Kyung Song Susan P. Ashdown
Article information:
To cite this document:
Youngsook Kim Hwa Kyung Song Susan P. Ashdown , (2016),"Women’s petite and regular body
measurements compared to current retail sizing conventions", International Journal of Clothing
Science and Technology, Vol. 28 Iss 1 pp. 47 - 64
Permanent link to this document:
http://dx.doi.org/10.1108/IJCST-07-2014-0081
Downloaded on: 10 November 2016, At: 18:27 (PT)
References: this document contains references to 29 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 199 times since 2016*
Users who downloaded this article also downloaded:
(2016),"Classifying male upper lateral somatotypes using space vectors", International Journal
of Clothing Science and Technology, Vol. 28 Iss 1 pp. 115-129 http://dx.doi.org/10.1108/
IJCST-03-2015-0043
(2016),"Pattern alteration of women’s suits based on ease distribution", International Journal
of Clothing Science and Technology, Vol. 28 Iss 2 pp. 201-215 http://dx.doi.org/10.1108/
IJCST-07-2015-0083
Access to this document was granted through an Emerald subscription provided by emerald-
srm:173961 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald
for Authors service information about how to choose which publication to write for and submission
guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as
well as providing an extensive range of online products and additional customer resources and
services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the
Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for
digital archive preservation.
*Related content and download information correct at time of download.
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
Women’s petite and regular body
measurements compared to
current retail sizing conventions
Youngsook Kim and Hwa Kyung Song
Department of Clothing and Textiles,
Kyung Hee University, Seoul, South Korea, and
Susan P. Ashdown
Department of Fiber Science and Apparel Design,
Cornell University, Ithaca, New York, USA
Abstract
Purpose –The purpose of this paper is to analyze petite women’s body size and figure and investigate
whether current petite sizing charts accurately reflect actual petit size women’s bodies. This study also
categorizes petite women’s body shapes and suggests primary body measurements as a base size for
each shape. The ultimate goal is to suggest fundamental body measurements for apparel companies to
modify and improve their sizing.
Design/methodology/approach –This study used data from SizeUSA data to compare body
measurements of 18-35-year-old petite women to regular women. The authors compared the results to
measurement differences between petite and regular sizing charts of 14 apparel companies. Then,
using the principal component analysis and cluster analysis, the authors classified petite women’s
body shapes. Body measurements for each body type are contrasted with the current petite sizing
charts, and then, the authors present differences as suggestions for modification and improvement of
petite sizing.
Findings –Industry sizing system do not generally represent average petite size women preciously
except for stature. Within the petite women, four body types were identified (top petite: 30.0 percent,
bottom petite: 30.8 percent, regular petite: 23.6 percent, and plus size: 15.4 percent). Of the four groups,
the ASTM D7878 generally represented the “top petite”sizing.
Originality/value –It is the first to analyze the industry petite sizing system utilizing population data
and focus petite sizing for women aged 18-35. The authors believe this study could draw attention of the
apparel industry, providing companies with ideas of how to improve their petite sizing for young women.
Keywords Body shape, Industry sizing, Petite size, SizeUSA data
Paper type Research paper
1. Introduction
Most ready-to-wear companies design clothing for consumers of average height and
weight ( Faust and Carrier, 2010). However, US national anthropometric survey
(SizeUSA) data shows that sizes other than those representing average height and
weight represent a significant percentage of the population (Textile Clothing
Technology Incorporated, 2004). For example, petite size females who are shorter than
5′4″represent 54 percent of the population ages 18-35. Therefore, females who do not
fall within these average statistical categories can have difficulty finding well-fitted
clothing in the marketplace (Shim and Kotsiopulos, 1990; Shim and Bickle, 1993; Jones
and Giddings, 2010; Park et al., 2009).
Apparel companies recognize the valuable potential for providing for special size
consumers based on certain physical characteristics like height (i.e. petite and tall).
Apparel producers like Gap and JCPenney do offer sizes that accommodate multiple
International Journal of Clothing
Science and Technology
Vol. 28 No. 1, 2016
pp. 47-64
© Emerald Group Publishing Limited
0955-6222
DOI 10.1108/IJCST-07-2014-0081
Received 4 July 2014
Revised 2 July 2015
Accepted 8 September 2015
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0955-6222.htm
47
Current retail
sizing
conventions
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
height categories like petite, average, and tall. Of all these special sizes, the petite
clothing market has been growing rapidly. The petite market is increasing annually by
more than 6 percent, and worldwide sales are expected to be USD57.7 billion by 2014
( Just-Style, 2008; Verdict Report, 2011). Still, in spite of this growth in demand,
approximately 50 percent of petite size women have reported it is hard to find petite
size clothing with the same level of fit as regular size clothing (NPD Group, 2012). While
petite women’s figures cannot be simply downscaled from a regular woman’s figure, a
general industry practice when producing petite size clothing is simply to scale down or
shorten regular size clothing as a solution (Liu, 2011; Huckabay, 1992; Jones and
Giddings, 2010). Fit problems may still appear and can be attributed to proportional
differences in clothing patterns between the average and the specialty sizes.
No study has yet compared the measurements in petite sizing systems to those of
regular sizing systems regularly used in the industry. If this comparison is conducted,
it would be possible to infer how apparel companies are actually defining petite size
women’s bodies and how petite size clothing is scaled down from regular size clothing.
The current study compares the petite and regular sizing systems. The study also
compares the results for the body measurements of petite size women to those for
regular size women using the US national anthropometric survey data (SizeUSA).
The results can help identify whether current retail sizing systems reflect actual petite
women’s bodies.
Beyond solving the issue regarding “downsizing”regular size measurements, we
assumed that if companies do target-specific body shapes within the petite population,
they will provide better fit for this sub-group if their sizing system covers the whole
range of petite women sizes. Therefore, we categorize petite size women’s body shapes
using statistical methods (principal component (PC) analysis and cluster analysis) and
suggest basic size measurements for petite sizing systems for each body shape group.
2. Literature review
2.1 Sizing standards for petite size misses women
The government has already provided voluntary apparel sizing standards to apparel
industry. The “petite”size classification was first introduced in the publication, Women’s
Measurements for Garment and Pattern Construction (O’Brien and Shelton, 1941).
Misses women were divided into three categories, namely, “short,”“regular,”and “tall.”
The “Short”designation was to be the first “petite”classification presented in
government documents.
The National Bureau of Standards (1958) then issued the Voluntary Product
Standard CS 215-58 titled, body measurement for the sizing of women’s patterns and
apparel, which classified women as “misses,”“women,”and “junior”as well as
identifying “short,”“regular,”and “tall”as groups.
National Bureau of Standards (1971) published PS 42-70 as a revised version of CS
215-58 and removed the “bust-hip”category and shifted the size designation for bust
girth. In terms of stature, PS 42-70 defined petite size as a woman whose height was
shorter than 5′2″. Petite size women were defined as women of shorter height (stature:
3½″, waist height: 2½″, and crotch height 1⅞″) and length (back waist length: ¾″, arm
length: 1½″, armscye depth: ⅛″) shorter than that of regular size women, but still
having the same girth and width measurements as the regular size.
In 1995, ASTM International superseded the National Bureau of Standards and
issued ASTM D5585, Standard Table of Body Measurements for Adult Female
Misses Figure Type, and in then 2011, this standard was updated based on SizeUSA
48
IJCST
28,1
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
data and current industry practices. The standard was only developed for misses’
regular size women.
In 2013, ASTM International newly developed the ASTM D7878 for Misses’petite
size women. Upon comparing ASTM D7878 to ASTM D5585, we found that the ASTM
International standard defined petite size women as a woman of a 3″shorter stature,
¾″shorter back waist length, 2″shorter leg length, and 1⅞″shorter arm length than
regular size women. In terms of girths, petite women were defined as having a ½″
larger waist girth and ½″narrower shoulder width than regular size women, while the
rest of their girth and width measurements stayed the same for the two standards.
2.2 Petite clothing fit
The government has provided voluntary sizing standards, but the apparel industry has
not adhered to these standards (Faust et al., 2006). Since each apparel company
has different target consumers with different body sizes and shapes, petite sizing
systems can vary for different companies (Barbaro, 2006). In general, apparel
manufacturers develop their own sizing systems by either copying already developed
size charts or adjusting those charts through trial and error, relying on customer
surveys or an analysis of sales and returned merchandise reports (Petrova, 2007).
Prior research studies have revealed that women not within the average height
range are less satisfied with their clothing fit than average size women (Shim and
Kotsiopulos, 1990; Shim and Bickle, 1993; Jones and Giddings, 2010). When Shim and
Bickle (1993) investigated differences in satisfaction with apparel fit for three size
groups ( petite, medium, and tall), the petite size group showed the least satisfaction
with 21 fit locations and two fit factors (sizing definition and ordering information).
Shim and Kotsiopulos (1990) also indicated that those groups who were members of the
average category for height and weight (e.g. tall women) were the least satisfied with
their ready-to-wear apparel. The NPD Group (2012) then found that approximately
50 percent of petite size women reported it was hard to find petite size clothing with the
same level of fit as regular size clothing.
One possible reason that petite consumers are not satisfied with their clothing fit is
that apparel companies have produced petite size clothing without sufficient
knowledge or a clear understanding of the petite woman’s body type. Petite women’s
figures cannot be simply downscaled from the regular women’s figure types, and yet
some apparel companies have decided to produce their petite size clothing by simply
scaling down or shortening hems of regular size clothing (Liu, 2011; Huckabay, 1992;
Jones and Giddings, 2010).
This lack of proportionate adjustment to actual petite women’s sizes increases the
probability that these customers will need to alter their petite-sized clothing after
purchasing it. Clothing companies need to understand petite size women’s figures more
precisely and thoroughly. Because the petite size proportion does have different
dimensions from sizes for children or smaller adults of regular height, petite clothing
should not simply become smaller in its aspects of length and circumferences.
Little research has addressed petite women’s body proportions and shapes (Kim,
1993; McRoberts, 2005). Kim (1993) compared petite women’s body proportions for
60 Asians and Caucasians. The results indicated that Caucasian women generally have
a longer lower torso than Asians do. McRoberts (2005) investigated figure-type
variations for 52 petite women. Various figure types, i.e., apple-shape or pear-shape,
were found, but none of the participants had the typical standard hourglass silhouette
as defined by the Voluntary Product Standard PS 42-70. No study has analyzed petite
49
Current retail
sizing
conventions
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
size women’s body proportions and figure types using the recent and larger amount of
anthropometric data now available.
2.3 Body shape classification methods
When female body shape categorization systems used in the research of apparel were
examined, three representative mathematical methods were found: first, drop
(difference) values of body circumference measurements; second, ratios of body
circumference measurements, and third, PC analysis and cluster analysis.
The first method utilizes the drop values between bust, waist, or hip circumference
measurements to identify body types. The International Organization for
Standardization (1991) published technical report, ISO/TR10652, standard sizing
systems for clothes, which offered guidelines to classify body shapes when creating a
sizing system based on anthropometric data. For example, for sizing systems for
women’s garments that cover the upper or whole body, the International Organization
for Standardization classified figure types based on the drop value between bust and
hip girth into an A type (9 cm), M type (4∼8 cm), and H type (3∼−4 cm or greater).
Several countries like Japan, Korea, Germany, and Hungary have utilized the first
method to create their own national sizing systems. For example, Korean Standard
Association (2009) classified female body types were based on the drop values between
bust and hip girth into an N type (3∼6 cm), A type (9∼21 cm), and an H type (−14∼3) for
garment-sizing systems for the upper body.
The second method for body shape analysis is based on the ratios of circumference
measurements for the bust, waist, high hip, abdomen, and hip. Simmons et al. (2004)
categorized nine whole body shapes (hourglass, bottom hourglass, top hourglass,
spoon, rectangle, diamond, oval, triangle, and inverted triangle) using five
circumference ratios. Lee et al. (2007) compared body types between US and Korean
women using the Simmons et al.’s (2004) method.
The first and second methods are relatively simple to use. However, they do not include
height and length measurements, so they are not appropriate for the current study
because vertical measurements are critical to analyze petite women’s figures adequately.
The third method used for body shape analysis applies both the PC analysis and
cluster analysis. PC analysis is used to reduce the number of measurement variables by
combining those measurements where the correlation is high into new composite
dimensions. Several prior studies have used PC scores as independent variables for a
cluster analysis when classifying body shapes.
Salusso-Deonier et al. (2006, 1985) utilized girths, lengths, and heights for PC
analysis to address the whole body shape. Most of the vertical measurements (heights
and lengths) belonged to one component, and most of the horizontal measurements
(girths) belonged to the other component. However, a cluster analysis could only
classify size-based groups like “tall and lean”and “tall and obese”because only the two
components were used as independent variables for the cluster analysis.
To extract a shape-based PC, Song and Ashdown (2011) used 15 drop values for
girths, front/back arcs, widths, and depths. The analysis extracted five shape-related
components (waist to top hip silhouette, top hip to hip silhouette, buttock prominence,
slope from front abdomen point to waist front point, and slope from front abdomen
point to front point at the hip level). Using a cluster analysis technique, “curvy”,“hip
tilt”, and “straight”shapes were identified.
The researchers in the current study concluded that the third method was
appropriate to analyze petite women’s body shapes because it used multiple variables,
50
IJCST
28,1
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
including both vertical and horizontal measurements. To avoid having vertical
measurements or horizontal measurements into either the first or the second PC,
we normalized measurements by stature and utilized “body measurements to stature
ratios”as independent variables for the PC analysis. By excluding the effect of stature
on other measurements, we also expected to extract PCs that would represent
distinctive body parts.
3. Research objectives
The purpose of this study is to provide information about body sizes and proportions of
petite women aged 18-35 and improve the petite size charts used for the apparel
industry. While the ASTM International (2013) developed a sizing for petite women
aged 18-55, we focussed women aged 18-35 because the body volume size such as bust
girth and hip girth is changed from over age 35 (Textile Clothing Technology
Incorporated, 2004). We assumed that there would be the problems when women aged
18-35 wear clothing generated from sizing charts of the industry sizing charts and
ASTM D7878 targeting women aged 18-55.
We analyzed women’s petite and regular body measurements and compared them
with the current retail sizing conventions. We expect the results from this study will
assist apparel companies in understanding petite size woman’s body size and figure
and determine whether their current sizing systems accurately reflect actual petite size
body. We further categorized petite size women’s body shapes and suggested primary
body measurements as a base size for each shape. Body measurements for each body
shape type are contrasted with the current petite sizing charts; differences are regarded
as suggestions for modification and improvement of petite sizing. The specific
objectives of the study are the following:
(1) Compare the industry petite size range to the size distribution of women ages
18-35.
(2) Compare the body measurements differences between petite and regular
industry size charts to those between petite and regular size women.
(3) Categorize petite size women’s body shapes and identify their body
characteristics.
(4) Compare body measurements for each body shape type with the current petite
sizing charts (ASTM and selected apparel companies).
4. Methodology
4.1 Comparison of industry petite size range to size distribution for women ages 18-35
4.1.1 Industry petite size range. From the report titled “The top 100 apparel specialty
stores ranked by industry sales”(Chain Store Guide, 2012), we selected 14
apparel companies that produce both petite and regular clothing and have sales over
USD500 million. We searched the petite sizing systems posted on their websites for
March, 2014.
All 14 companies included bust girth, waist girth, and hip girth in their sizing
systems. Eight companies additionally listed leg length, and seven companies posted
arm length. Waist to hip length, thigh length, and total crotch length were listed by
only one company, so we excluded the three measurements from the analysis. In terms
of stature, 12 out of 14 companies listed a standard below their charts, so we included
51
Current retail
sizing
conventions
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
stature in the analysis. Finally, we investigated the range of the body measurements
across size range (measurements for minimum and maximum) for six measurements
(bust girth, waist girth, hip girth, leg length, arm length, and stature).
4.1.2 Size distribution for women ages 18-35. We utilized the data set from the
SizeUSA project, a national sizing survey using a 3D body scanner (Textile Clothing
Technology Incorporated, 2004). We selected 2,981 women ages 18-35. Then, we
removed 267 women (9.0 percent) who had stature taller than 5′8″, considered to be in
the tall size category as defined by Jones and Giddings (2010). Using data for the 2,714
women, we calculated the percentiles for six measurements and compared those results
to the industry size ranges.
4.2 Comparison of body measurement differences for petite and regular size women
to measurements for petite and regular sizing systems
4.2.1 Comparison of body measurements of petite to regular size women. To compare
the body measurements for petite size women to those of regular size women, we
divided the 2,714 women into petite and regular size categories. We defined petite size
as a woman shorter than 5′4″(Rasband and Liechty, 2006; Faust and Carrier, 2010).
As a result, 1,618 women were selected for the “petite”group, and 1,096 women were
selected for the “regular”group.
To compare body measurements for the two groups, we conducted a t-test based on
a 95 percent confidence level to compare the mean scores for 26 body measurements
(seven heights, five lengths, 11 girths, and three widths) and two drop values
(bust-waist girth and hip-waist girth) considered critical for proper clothing fit.
4.2.2 Comparison of body measurements for petite sizes to regular industry sizing
systems. Based on the sizing charts posted on 14 apparel company websites, we
calculated six measurement (stature, bust girth, waist girth, hip girth, leg length, and
arm length) differences between the petite and regular sizing systems using the same
size code. We compared these to measurement differences found between the petite and
regular groups for the SizeUSA data.
4.3 Petite women’s body classification method
To classify petite women’s body shapes, we first conducted a PC analysis (varimax
rotation method) using the statistical package SPSS 17.0. We selected 14 measurements
(five girths, three widths, and six lengths) critical for developing pattern blocks and
sizing systems (Table I). We divided each body measurement by stature, and utilized
14 body measurements by stature ratios for independent variables for a PC analysis.
The PC scores were used for independent variables of the following cluster analysis.
To classify the body shapes, we conducted two statistical analyses: first, X-means
cluster analysis using R program, second, K-means cluster analysis using the statistical
package SPSS 17.0. As the two methods can provide different intuition for structure of
data, we considered both methods to choose the number of clusters.
First, we programmed X-means using R program since there is no X-means in SPSS
program. X-means is a hierarchical model selection method. It performs two-means
clustering and checks by partial Bayesian information criterion value whether each
cluster should be further split into two sub-clusters. The final number of clusters were
automatically suggested by the program.
As a different clustering method, the K-means cluster analysis was conducted
according to the algorithm implemented in SPSS program. When we specified the
52
IJCST
28,1
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
number of clusters (K) to the program, the program first randomly assigned an initial
set of cluster centers. Then, the program found each woman to the nearest cluster
center by comparing Euclidean distances from each woman’s points to each
cluster centers. For each of Kclusters, the program found cluster centroid and update
cluster center location to centroid. The program repeated this procedure until the K-
means algorithm terminated when the centroids no longer change. We experimented
with different clusters, dividing the petite size women (n¼1,618) into two to five
groups. The researchers in the current study decided the final number of clusters based
on the significant differences between the clusters and a reasonable number of clusters
for further study such as the development of the sizing systems. Cluster mean of each
component score was computed and compared through Duncan’s multiple range tests
of the ANOVA post hoc analysis.
The goal of this study was not to look for the best separation of the data into distinct
clusters, but to determine factors and define body measurements for our solution to
sizing issues, resulting in the best answer for informing and creating sizing systems.
Considering the goal of the study, we determined the final number of clusters by
comparing the results of both statistical methods.
4.4 Comparison of body measurements for four petite groups to ASTM and industry
sizing charts
We calculated mean values for the 12 primary measurements of four body shape groups.
These were used for base size measurements for the sizing systems for these groups.
We compared these values with mid-sizes in ASTM D7878 and averages of 14 industry
sizing systems using one-way ANOVA analysis.
Measurement
location
Definitions from SizeUSA study (Textile Clothing Technology Incorporated,
2004)
Bust girth Measure bust circumference horizontally around the body under the arms,
across the nipples, and parallel to floor
Waist girth Measure body circumference at the smallest point of waist when viewed
from the side
High hip girth Measure high hip circumference of the body at high-hip level, approximately
3 inches below waist level and parallel to floor
Hip girth Measure circumference of hip measured around fullest part of buttocks
(at maximum posterior prominence of buttocks)
Max-thigh girth Measure circumference of upper leg 1 inch below crotch
Cross chest width Measure across chest front from armscye to armscye at the anterior
axillary fold point level
Cross back width Measure across back from armscye to armscye at the posterior axillary
fold point level
Across shoulder Measure across back from one shoulder point to the other on approximately
a 45-degree line
Front waist length Measure from center front neck base line to center front waist level
Back waist length Measure from cervical following contour of the spinal column to center
back waist level
Pelvic length Measure from waist circumference level vertically down to crotch level
Crotch height Standing erect without shoes with feet slightly apart, measure from
crotch straight down to soles of feet
Knee height Measure from knee circumference level vertically down to floor
Arm length Measure from shoulder point along the outside of the arm to wrist joint
Table I.
SizeUSA
measurement
locations used in
principal component
analysis
53
Current retail
sizing
conventions
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
Then, we additionally calculated the mean differences between petite and regular
sizes for four body shape groups, and compared them to those of the ASTM standards
(ASTM D5585 (regular size)-ASTM D7878 (petite size)) and industry sizing (regular
size-petite size).
5. Results and discussions
5.1 Industry petite size range vs size distribution for women
In Table II, industry size ranges are shown using arrows and the number of companies
for each size range is shown in parentheses next to each arrow. In terms of stature, the
industry petite size category generally fit half the women, and three out of 11
companies set maximum stature at 5′3″, while eight companies defined it as 5′4″, the
value at the 50th percentile of SizeUSA data. The industry standard of 5′4″was also
consistent in the standards used in the prior research studies ( Faust and Carrier, 2010;
Rasband and Liechty, 2006).
In both petite and regular size charts, bust girth, waist girth, and hip girth
measurements were broadly distributed throughout the SizeUSA data. When
considered as a whole, the apparel companies did not set differences for size
distribution between the two charts.
None of the companies assigned size intervals for inseam length between sizes and set
their inseam length at 29″(50th percentile), 29.5″or 30″. Industry size for arm length
was distributed above the 50th percentile for SizeUSA data.
Notes: The range of the sizing system for this variable; ( ), the number of companies with
the size range indicated by the arrow
Table II.
Size range of
industry petite sizing
systems and size
distribution of
women ages 18-35
from SizeUSA data
54
IJCST
28,1
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
5.2 Body measurements of petite vs regular size women
We compared 26 measurements and two drop values for petite size women to those of
regular size women in the SizeUSA data (see Table III). The petite group showed
significantly lower mean values than the regular group for all the vertical
measurements. The petite group had a 3.7″shorter stature, 2.16″lower crotch
height, and 1.26″lower knee height on average than the regular size group ( po0.05).
The petite group also had significantly shorter lengths than the regular group by 0.14″
at shoulder length, 0.82″at the back waist, and 1.52″shorter for arm length ( po0.05).
From the results, we recommend a degree of reduction in measurements for petite
sizes. Petite women were 0.82″shorter at back waist length, 0.5″(¼2.66 –2.16) at waist
Petite
(n¼1,618)
Regular
(n¼1,096)
Body measurements Mean SD Mean SD
Mean difference
(regular-petite) tSig.
Height
Stature 62.06 1.69 65.76 0.93 3.70 −73.44 0.00**
Cervical height 55.40 1.80 58.82 1.19 3.42 −59.73 0.00**
Waist height 38.08 1.88 40.74 1.58 2.66 −39.97 0.00**
Hip height 30.21 2.08 32.24 1.95 2.03 −25.92 0.00**
Abdomen height 35.39 2.05 37.91 1.88 2.52 −33.03 0.00**
Crotch height 27.97 1.41 30.14 1.16 2.16 −43.67 0.00**
Knee height 16.79 0.91 18.06 0.76 1.26 −39.19 0.00**
Length
Shoulder length 4.90 0.53 5.05 0.54 0.14 −6.89 0.00**
Front waist length 14.35 1.62 14.74 1.61 0.40 −6.27 0.00**
Back waist length 16.79 1.09 17.60 1.17 0.82 −18.38 0.00**
Arm length 20.25 1.23 21.77 1.09 1.52 −33.98 0.00**
Center back neck to wrist 27.51 1.44 29.24 1.27 1.73 −33.07 0.00**
Girth
Neck base 14.60 1.49 14.81 1.13 0.21 −4.02 0.00**
Bust 39.09 4.91 39.95 5.08 0.86 −4.41 0.00**
Waist 32.60 5.08 33.32 5.33 0.72 −3.57 0.00**
Hip 40.61 4.81 41.85 5.02 1.24 −6.46 0.00**
Max-thigh girth 23.96 2.89 24.85 3.05 0.89 −7.73 0.00**
Mid-thigh girth 19.76 2.44 20.39 2.57 0.64 −1.56 0.12
Knee 14.90 1.39 15.39 1.44 0.49 −8.86 0.00**
Calf 14.61 1.50 15.04 1.55 0.43 −7.19 0.00**
Armscye 15.78 1.86 16.31 1.84 0.53 −7.34 0.00**
Upper arm 11.55 1.93 11.79 2.03 0.24 −3.13 0.00**
Elbow 9.93 1.24 10.23 1.18 0.30 −6.30 0.00**
Width
Across shoulder 14.53 1.24 14.94 1.24 0.42 −8.55 0.00**
Cross chest width 14.55 1.84 14.72 1.89 0.16 −2.26 0.02*
Cross back width 13.85 1.39 14.42 1.40 0.56 −10.30 0.00**
Drop
Bust girth-waist girth 6.491 1.804 6.628 1.902 0.137 −1.880 0.06
Hip girth-waist girth 7.729 2.151 8.151 2.161 0.422 −5.006 0.00**
Notes: *po0.05; **po0.01
Table III.
Mean and standard
deviation for petite
and regular size
groups for 28
measurements
(unit: inch)
55
Current retail
sizing
conventions
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
to crotch height, 2.16″at inseam length, 0.9″(¼2.16 –1.26) at crotch to knee height,
1.26″at knee length, and 1.73″at arm length.
Also, there was significant difference between the two groups for most girths and
widths. The petite group also had significantly smaller girths than the regular group by
0.86″at bust, 0.72″at waist, 1.24″at hip, and 0.24″at upper arm ( po0.05). The t-test on
the hip-waist drop showed that the petite group had a straighter body shape between
waist and hip ( po0.05). As shown above, petite size should be scaled down from
regular size differently at each location.
5.3 Measurement differences for petite and regular size women vs differences between
petite and regular sizing systems
Table IV shows the mean differences for six measurements between petite and regular
women from the SizeUSA data and those for the petite and regular sizing systems of 14
companies. There were significant differences at bust girth (0.9″), waist girth (0.7″), and
hip girth (1.2″) between the petite and the regular groups in the SizeUSA data. Six (A-F
in Table IV) companies assigned no difference between the two sizing systems, and
four companies (G-I) reduced the three girths by 0.5″. Two companies ( J-K) reduced
them by 1″from their regular chart measurements on the petite chart, the similar result
as for the SizeUSA data. The final three companies (L-N) set different entries for bust,
waist, and hip measurements. These companies might be targeting specific petite
women with different bust-waist-hip proportions, or they might misunderstand petite
women’s actual body sizes.
For leg length, six companies reduced 3″from their regular size, but the difference
between the two groups for the SizeUSA data were 2.2″. Further, (H) company assigned
2.5″and (L) company did 2.75″, so apparel companies tended to reduce more inseam
length than actually needed.
For arm length, the difference between the two groups from SizeUSA data were 1.7″;
however, three companies (A, E, and L) reduced the length by 1.25″. Therefore, we
estimate petite size consumers may find their sleeve lengths to be longer than their
actual arm length.
5.4 PCs extracted for petite women’s body classification
Three PCs were extracted (see Table V). Three PCs indicated a 77.3 percent variation
for the 14 variables. The first PC had high loadings with girth and width
measurements, while the second PC had a high correlation of upper body length and
pelvic length. The last PC had high loadings for lower body height and arm length.
Therefore, we interpreted PC1 as “body volume,”PC2 as “torso length,”and PC3 as
“limb length.”
5.5 Classification of petite size women’s body shapes
To categorize the body shapes, we conducted two statistical analyses: first, X-means
cluster analysis using R program; second, K-means cluster analysis using the statistical
package SPSS 17.0. The X-means clustering method determined that two cluster should
not be further split, so suggested that two-group model is appropriate. The results are
as shown in Table VI. When two groups were categorized, they showed significant
differences in three PC scores, but the differences in PC1 scores (body volume) was
small. Group 1 relatively had the short torso length and long limbs, while Group 2 has
the reverse shape of Group 1.
56
IJCST
28,1
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
Measurement differences for petite and regular sizing systems used in apparel companies
(A) (B) (C) (D) (E) (F) (G) (H) (I) (J) (K) (L) (M) (N)
Measurements
Mean differences
between petite
and regular size
groups in
SizeUSA data
Lands’
End
Eddie
Bauer
Coldwater
Creek Express
L.L.
Bean Gap Talbots
Old
Navy
Banana
Republic
Ralph
Lauren
J.
Crew
Ann
Taylor
New
York &
Company
Jones
New
York
Bust girth 0.9** 0 0 0 0 0 0 0.5 0.5 0.5 1 1 1 1 0.5
Waist girth 0.7** 0 0 0 0 0 0 0.5 0.5 0.5 1 1 0.5 1 1
Hip girth 1.2** 0 0 0 0 0 0 0.5 0.5 0.5 1 1 1 0 1
Crotch height
(inseam length) 2.2** 3 3 3 3 3 2.5 3 2.75
Center back neck
to wrist (arm
length) 1.7** 1.25 1.5 1.25 1.5 1.25
Stature 3.7** 4 4.5 4 3 5 6 7 5 5 3
Note: **po0.01
Table IV.
Range of differences
for petite and regular
groups in SizeUSA
data compared to
actual differences
between petite and
regular sizing
systems used in
apparel companies
(unit: inches)
57
Current retail
sizing
conventions
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
K-means cluster method did not suggest the optimal number of clusters, so we should judge
which number of the clusters would be appropriate for our study. After dividing petite size
women into two to five groups, we concluded that four groups were the most appropriate
designation to represent petite women’s body shapes accurately for the following reasons.
When two groups were categorized, they did not show significant differences in PC1
scores (body volume). When three groups were categorized, Groups 1 and 2 had similar
PC1 scores, even though they were significantly different ( po0.05). In the five-group
model, Group 1 represented only about 5 percent of the total data.
In the four-group model, the data relatively evenly distributed (Group 1: 30.0 percent,
Group 2 was 30.8 percent, Group 3 was 23.6 percent, and Group 4 was 15.4 percent).
Table VII shows that the overall Ffor the one-way ANOVA was statistically different
(po0.05). Front silhouettes and profile sketches of each body shape group are shown
in Figure 1 for comparison. The main body shape characteristics of each body shape
group were:
•Group 1 (top petite) had the shortest torso and relatively average limbs.
•Group 2 (bottom petite) had the shortest limbs and an average torso length.
•Group 3 (regular petite) had a relatively longer torso and long limbs.
•Group 4 (plus size) had larger body volume.
Principal component
Components Variables 1 2 3
Body volume
(girths and widths)
Bust girth/height 0.963 0.010 −0.024
Waist girth/height 0.958 0.071 −0.060
High hip girth/height 0.957 0.051 −0.092
Hip girth/height 0.948 −0.111 −0.075
Across shoulder/height 0.890 0.053 −0.076
Max-thigh girth/height 0.881 −0.152 −0.014
Cross back width/height 0.785 0.094 −0.083
Cross chest width/height 0.670 −0.027 0.052
Torso length
(waist and pelvic length)
Waist length, front/height 0.332 0.829 −0.124
Waist length, back/height −0.008 0.807 −0.269
Pelvic length (waist-crotch height)/height 0.471 −0.801 −0.104
Limb length
(arm and leg length)
Knee height/height −0.032 −0.108 0.848
Crotch height/height −0.297 −0.144 0.830
Arm length, shoulder to wrist/height 0.105 −0.036 0.700
Total rotation sums of squared loadings 6.725 2.070 2.029
% of variance 48.039 14.787 14.496
Cumulative % 48.039 62.826 77.322
Table V.
Rotated component
matrix for principal
component analysis
of 14 body
measurements by
stature ratios
Body shape group
12
Components (n¼740) (n¼873) tSig.
Body volume −0.083 0.070 9.442 0.002
Torso length −0.721 0.611 1,271.676 0.000
Limb length 0.490 −0.416 412.469 0.000
Table VI.
Mean for three
principal components
for two shape
groups from X-mean
analysis
58
IJCST
28,1
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
The X-means analysis generated separation of the data into distinct two clusters, but
we did not use clustering to look for the best separation possible of the data, but to
determine factors and define body measurements for our solution to sizing issues.
A number of petite size women who shop petite size for blouses, jackets, and other tops,
but buy their jeans and pants from regular sizing systems, and that the reverse is true.
We also know that some women shop only in petite, and in addition that there is a
demand for plus size petite (e.g. Jessica London and Lane Bryant). The two-cluster
solution from X-means analysis does not address the needs of the overall petite women
or of the plus size petite.
62”(5’2”)
16”
Group 1 (Top Petite) Group 2 (Bottom Petite) Group 3 (Regular Petite) Group 4 (Plus Petite)
Group 1 (Top Petite) Group 2 (Bottom Petite) Group 3 (Regular Petite) Group 4 (Plus Petite)
62”(5’2”)
20 ”
3430 ”
14
39 ”
1838 ”
14
30 ”
38
19 ”
3820 ”
3420 ”
18
33 ”
14
40 ”
1448 ”
12
40 ”
12
28 ”
34
10 ”
34
17 ”
1417 ”
1216 ”
12
11 ”
12
9 ”
349 ”
18
27 ”
316 27 ”
14
28 ”
38
Figure 1.
Front and side
views of four
body shapes
Body shape group
Components
1
(n¼486)
2
(n¼497)
3
(n¼381)
4
(n¼249) FSig.
Body volume −0.436
c
−0.465
c
0.105
b
1.618
a
607.666 0.000
Torso length −0.873
d
0.305
b
1.015
a
−0.457
c
615.203 0.000
Limb length 0.444
b
−0.830
d
0.724
a
−0.318
c
374.826 0.000
Note: aWbWc
Table VII.
Mean for three
principal components
for four shape
groups from
K-means analysis
59
Current retail
sizing
conventions
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
Therefore, we chose four-group model from K-means cluster analysis. No study has
analyzed actual petite size women’s body proportions based on anthropometric
data to validate these different configurations. Our classification results show why
a number of petite women should buy top and bottom clothing in different size
categories, namely, that almost 60 percent of petite women (Group 1 (top petite, 30
percent), Group 2 (bottom petite, 30.8 percent)) did have different top and bottom
proportions. Apparel companies should clearly acknowledge these customers’
needs because when petite women wish to purchase dress or coat items, both the top
and the bottom must fit well.
We also learned that 15.4 percent of the petite women can be classified as the “plus
size”group. However, when we searched the petite size producers, only a few firms, such
as Talbots, Lane Bryant, and Lands’End, actually produced clothing for plus size
women. Even though the plus size petite market is not a small one, apparel firms remain
rather shortsighted about this market. The apparel industry should acknowledge the
growing potential of the plus size petite clothing market and address it.
5.6 Body measurement of four petite sizing charts vs ASTM and industry sizing charts
Table VIII presented the results of comparison of mean values for the primary
measurements of four body shape groups to values of mid-sizes in ASTM D7878 and
averages of 14 industry sizing systems from one-way ANOVA analysis.
ASTM D7878 represented girth measurements of the industry sizing systems
almost perfectly. Of our four shape groups, the ASTM D7878 generally represented
“top petite”sizing. However, the other shape group sizing systems were different from
ASTM D7878.
Table IX presented the mean differences between petite and regular sizes for four
body shape groups, the ASTM standards (ASTM D5585 (regular size)-ASTM D7878
(petite size)) and industry sizing (regular size-petite size). Even when apparel companies
have different regular sizing systems, they can develop their petite sizing systems
further by deducting the values shown in Table VIII. The differences gathered from the
ASTM standards were generally similar to those for the “top petite”group. The rest of
the shape groups were different from the ASTM standards.
6. Significances, limitations, and future research opportunities
This study is significant in that it is the first to analyze the clothing industry petite
sizing system utilizing population data. This study focusses on petite sizing for women
aged 18-35, while ASTM D7878 was developed for women aged 18-55. We also
categorized petite size women’s body shapes using PC analysis and cluster analysis
and suggested new primary base size body measurements for each shape group.
The point of the current study is not to simply “downsize”regular measurements, but
create new sizing standards that more precisely fit the actual needs of petite women in
search of clothes that fit them well by providing sizing standards targeting specific
body shapes within the petite population. If companies do target-specific body shapes
within the petite population, they will provide better fit for this sub-group. The results
of this study can assist apparel companies in revising their petite sizing systems and
thus better understand petite size women’s bodies and figure types and improve the fit
of their petite size clothing.
The limitation of this study was basing the analysis of charts by only 14 companies.
Therefore, the results may not be representative of all actual petite sizing systems in
60
IJCST
28,1
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
Petite groups in current study
Measurements
ASTM D7878
(average of sizes 8 and 10) Top petite Bottom petite Regular petite Plus petite Industry FSig.
Height and length
Stature 62 1/2
a
62 1/2
a
61 5/8
a
62 1/16
a
61 15/16
a
62
a
13.455 0.000**
Back waist length 15 3/8
d
16
c,d
17 1/4
a,b
17 1/2
a
16 1/2
b,c
–189.621 0.000**
Waist to hip length 7 5/8 (curvy)
b,c
7 1/2 (straight)
b,c
9
a
8
a,b
7
b,c
6 5/8
c
–154.534 0.000**
Waist to crotch length 10 (curvy)
b,c
10 5/8
b
9 5/8
c,d
9
d
11 1/2
a
–369.653 0.000**
9 7/8 (straight)
c
10 5/8
b
9 5/8
c
9
c
11 1/2
a
–369.679 0.000**
Crotch height 28 1/2
a,b
28 3/4
a,b
27 3/16
c
28 3/8
b
27 1/4
c
29 1/2
a
123.284 0.000**
Knee height 16 1/2
b,c
17 1/4
a
16 1/4
c
17 1/8
a,b
16 5/8
a,b,c
–147.360 0.000**
Center back neck to wrist length 28 5/8
b
27 3/4
b
26 1/2
c
28 1/16
b
28 1/4
b
30 1/8
a
114.230 0.000**
Shoulder to wrist length 21 1/8
a
20 3/4
a
19 3/8
b
20 3/4
a
20 1/8
a,b
–113.723 0.000**
Girth and width
Bust girth 36 3/4
b
37 1/4
b
36 3/4
b
39 1/2
b
46 1/2
a
36 1/8
b
292.731 0.000**
Waist girth 29 (curvy)
c
30 1/4
b,c
30 3/8
b,c
33 1/4
b
40 1/2
a
29
c
334.274 0.000**
30 1/2 (straight)
b,c
30 1/4
b,c
30 3/8
b,c
33 1/4
b
40 1/2
a
29
c
334.008 0.000**
Hip girth 39 3/4 (curvy), 39 (straight)
b
39 1/8
b
38 1/4
b
40 1/4
b
48 1/2
a
39
b
342.446 0.000**
Across shoulder width 15 1/6
b
14 1/8
c
14 1/16
c
14 5/8
b,c
16 3/16
a
–217.813 0.000**
Note: aWbWcWd
Table VIII.
Mid-size ASTM
D7878, averages of
four petite groups
compared to
averages of 14
apparel companies
61
Current retail
sizing
conventions
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
the garment industry. In the future, researchers need to obtain spec sheets that
include measurements from more companies to validate the results of this
study. While this study analyzed only a specific population of US women ages
18-35, future research needs to analyze older women’s petite sizes, petite women of
different ethnicities, and even men’s petite sizing. This study can be considered as the
first step in revising the entire petite sizing for all those in need of that kind
of clothing.
References
ASTM International (2013), ASTM D7878-13: Standard Tables of Body Measurements
for Adult Female Misses Petite Figure Type, ASTM International, West Conshohocken, PA.
Barbaro, M. (2006), “Where’s the petite department? Going the way of the petticoat”,The
New York Times, 28 May, p. A1(L).
Chain Store Guide (2012), Market Research: The Top 100 Apparel Specialty Companies, Chain
Store Guide, Tampa, FL.
Faust, M.E. and Carrier, S. (2010), “Women’s wear sizing: a new labelling system”,Journal of
Fashion Marketing and Management, Vol. 14 No. 1, pp. 88-126.
Faust, M.E., Carrier, S. and Baptiste, P. (2006), “Variations in Canadian women’s ready-to-
wear standard sizes”,Journal of Fashion Marketing and Management, Vol. 10 No. 1,
pp. 71-83.
Huckabay, D.A. (1992), “Perceived body cathexis and garment fit and style proportion problems
of petite women”, unpublished master’s thesis, Virginia Polytechnic Institute and State
University, Blacksburg, VA.
International Organization for Standardization (1991), ISO/TR 10652: Standard Sizing Systems
for Clothes, International Organization for Standardization, Geneva.
Petite groups in current study
Measurements ASTM
Top
petite
Bottom
petite
Regular
petite
Plus
petite Industry
Height and length
Stature 3 3 1/4** 4** 3 5/8** 3 7/8** 4 3/4
Back waist length 3/4 5/8* 5/8 1 1/8 7/8 –
Waist to hip length 3/8 (curvy)
1/2 (straight) 3/8 1* 5/8 3/8 –
Waist to crotch length 0 (curvy)
1/8 (straight) 1/4** 7/8 3/8 3/4 –
Crotch height 2 2 1/8** 2 3/8** 1 3/4* 2 1/4** 2 7/8
Knee height 1 3/8 1 1/4** 1 3/8** 1* 1 1/4 –
Center neck to wrist length 2 1/8 1 5/8 1 7/8** 1 3/8* 2 1 3/8
Shoulder to wrist length 1 7/8 1 1/2 1 5/8** 1 1/8* 1 5/8 –
Girth and width
Bust girth 0 5/8* 1** 1/2 2 1/2 3/8
Waist girth −1/2 1/2 5/8 5/8 2 1/4 3/8
Hip girth 0 3/4 1 1/2 1 2 1/2 3/8
Across shoulder width 1/2 1/2** 3/8** 1/2 3/4 –
Notes: *po0.05; **po0.01
Table IX.
Measurement
differences for
ASTM D5585 and
ASTM D7878,
regular and petite
groups in SizeUSA
data, and average
differences between
petite and regular
sizing charts in 14
apparel industries
62
IJCST
28,1
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
Jones, M.R. and Giddings, V.L. (2010), “Tall women’s satisfaction with the fit and style
of tall women’s clothing”,Journal of Fashion Marketing and Management, Vol. 14 No. 1,
pp. 58-71.
Just-Style (2008), Global Market Review of Outsize Clothing –Forecasts to 2014, Bromsgrove.
Kim, S.H. (1993), “Analysis of petite-sized women’s body proportions related to garment-fit”,
doctoral dissertation, University of Minnesota, Minneapolis, MN.
Korean Standard Association (2009), KS K 0051: Sizing Systems for Female Adult’s Garments,
Korean Standard Association, Seoul.
Lee, J.Y., Istook, C.L., Nam, Y.J. and Park, S.M. (2007), “Comparison of body shape between USA
and Korean women”,International Journal of Clothing Science and Technology, Vol. 19
No. 5, pp. 374-391.
Liu, N.N. (2011), “Women struggle with being a size smaller than small”, available at: www.
nwasianweekly.com/2011/05/women-struggle-with-being-a-size-smaller-than-small/
(accessed January 31, 2014).
McRoberts, L. (2005), “Petite women: fit and body shape analysis”, unpublished master’s thesis,
Louisiana State University, Baton Rouge, LA.
National Bureau of Standards (1958), Voluntary Product Standard CS 215-58: Body
Measurements for the Sizing of Women’s Patterns and Apparel, US Department of
Commerce, Washington, DC.
National Bureau of Standards (1971), Voluntary Product Standard PS 42-70: Body Measurements
for the Sizing of Women’s Patterns and Apparel, US Department of Commerce,
Washington, DC.
NPD Group (2012), Women Special Sizes Reports: Tops and Bottoms and Bra and Panty, NPD
Group, Port Washington, NY.
O’Brien, R. and Shelton, W.C. (1941), Women’s Measurements for Garment and Pattern
Construction, US Department of Agriculture, Washington, DC.
Park, J., Nam, Y., Choi, K., Lee, Y. and Lee, K. (2009), “Apparel consumers’body type and their
shopping characteristics”,Journal of Fashion Marketing and Management, Vol. 13 No. 3,
pp. 372-393.
Petrova, A. (2007), “Creating sizing systems”, in Ashdown, S.P. (Ed.), Sizing in Clothing,
Woodhead Publishing Limited, Cambridge, pp. 57-87.
Rasband, J.A. and Liechty, E.G. (2006), Fabulous Fit, Fairchild Publications, New York, NY.
Salusso-Deonier, C.J., Borkowski, J.J., Reich, N. and Goldsberry, E. (2006), “An alternative
approach to sizing apparel for women 55 and older”,Clothing and Textiles Research
Journal, Vol. 24 No. 2, pp. 96-111.
Salusso-Deonier, C.J., Delong, M.R., Martin, F.B. and Krohn, K.R. (1985), “A multivariate method
of classifying body form variation for sizing women’s apparel”,Clothing and Textiles
Research Journal, Vol. 4 No. 1, pp. 38-45.
Shim, S. and Bickle, M.C. (1993), “Women 55 years and older as catalog shoppers: satisfaction
with apparel fit and catalog attributes”,Clothing and Textiles Research Journal, Vol. 11
No. 4, pp. 53-64.
Shim, S. and Kotsiopulos, A. (1990), “Women’s physical size, body-cathexis, and shopping for
apparel”,Perceptual and Motor Skills, Vol. 71 No. 3, pp. 1031-1042.
Simmons, K.P., Istook, C.L. and Devarajan, P. (2004), “Female figure identification technique
(FFIT) for apparel, part II: development of shape sorting software”,Journal of Textile and
Apparel, Technology and Management, Vol. 4 No. 1, pp. 1-15.
63
Current retail
sizing
conventions
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)
Song, H.K. and Ashdown, S.P. (2011), “Categorization of lower body shapes for adult
females based on multiple view analysis”,Textile Research Journal, Vol. 81 No. 9,
pp. 914-931.
Textile Clothing Technology Incorporated (2004), SizeUSA, The National Sizing Survey –
Women, Textile Clothing Technology Incorporated, Cary, NC.
Verdict Report (2011), UK Niche Clothing Segments 2011, Verdict Publications, London.
Corresponding author
Hwa Kyung Song can be contacted at: hksong@khu.ac.kr
For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com
64
IJCST
28,1
Downloaded by HONGIK UNIVERSITY At 18:27 10 November 2016 (PT)