Women and Girls in Science, Technology,
Engineering, and Mathematics
cience, technology, engineering, and ~arhematics (STEM) are
widely regarded as critical to the ?~tto~al economy. Concern
about America's ability to be competitive m the glob~! e~onomy
has led to a number of calls to action to strengthen the p1pelme tnto
these fields (National Academy of Sciences, Committee on Science,
Engineering & Public Policy, 2007; U.S. Government AccountabilIty Office, 2006; U.S. Department of Ed~catio~'. 200?). Expandmg
and developing the STEM workforce IS a critical tssue for government, industry leaders, and educators. Despite the tremendous
gains that girls and women have made in education and the workforce during the past 50 years, progress has been uneven, and certain scientific and engineering disciplines remain overwhelminglv
male. This report addresses why there are stdl so few women i~
certain scientific and engineering fields and provides recommendations to increase the number of women in these fields.
. Th~ . National Science Foundation estimates that about
ftve mtllJOn ~eople work dtrectly in science, engineering, and
technology- JUSt over 4 percent of the workforce. 1 This relatively
~mall g~oup of workers is considered to be critical to economic
~nno~atiOn and productivity. Workers in science and engineermg fields tend to be well paid and enjoy better job secunty than
do other workers. Workforce projections for 2018 b
h US
e .' ·
10 astest-growtng
occupations t at reqUJre at least a b h 1 ,
significant scientific or mathemat· lac ~ ~r s degree wdl reqUJre
1ca trammg Man
engmeenng occupations
· l
Y sctence an
are pre 1cted to g
age rate for all occupations
row aster t an the averbe in engineering- and co' an somle of the largest increases will
mputer-re ated f ld
f' ld .
. h
women currently hold onete s- 1e s m wh1c
Wright, 2009; National ScieqnuarBter odr fewer positions (Lacey &
ce oar 2010)
Attractmg and retaining m
. ·
will maximize innovation c or.e ~omen 10 the STEM workforce
, reat1v1ry and
t1sts an engmeers are work'
competltlveness. Scienmg to solv
c h a 11 enges of our time-find·
e some of the most vexing
mg cures f d'
rna Iana, tackling global w
or tseases like cancer and
armtng pr 'd·
. k'
mg people with clean
. .
stan d mg t e ongms of the u ·
energy sources and undernt verse En .
· gmeers design ' many of the


Definition of Sdence,
Technology, Engineering,
and Mathematics (STEM)
STEM is defined tn many ways (for
example, see U.S. government definitions
at http1/
.pdf). II" this report the term "STEM"
refers to the phystcal, b1olog1Cal, and
agricultural setences; computer and
tnformation sciences; engtneering
and engineenng technologtes; and
mathematiCS. The social and behavioral
sciences, such as psychology and
economics. are not .ncluded, nor are
health workers. such as doctors and
nurses. College and university STEM
faculty are 1ncluded when possible, but
h1gh school teachers in STEM subjects are
not While all of these workers are part
of the larger soent1fic and eng1neering
workforce, the1r exclusion is based on
the availabtltty of data. In thts report the
terms "STEM.~ "soence, technology,
engtneenng, and mathemattcs,n and
•soentiftc and eng1neenng fields" are
used mterchangeably.



Women and Girls in Science, Technology, Engineering, and Mathematics

things we use da1ly-buildings, bridges, computers, cars, wheelchairs, and X-ray machines. When women are not involved in
the destgn of these products, needs and desires unique to women
ma) be overlooked. For example, "some early voice-recognition
S}Stems were calibrated to typical male voices. As a result, women's vo1ces were literally unheard .... Similar cases are found in
nMn) other industries. For mstance, a predominantly male gro up
ot engmcers ratlored the first generation of automotive airbags
to adult male bodies, resulting in avoidable deaths for women
and children" (Margolis & F1sher, 2002, pp. 2-3). With a more
d1\erse workforce, scientific and technological products, services,
and solutions are likely to be better designed and more likely to
represent all u ers.
The opportunity to pursue a career in science, technology,
engineering, and mathematics is also a matter of pay equity. Occupational segregation accounts for the majority of the wage gap
(AAUW Educational Foundation, 2007), and although women
till earn less than men earn in science and engineering fields, as
they do on average in the overall workforce, women in science
and engineering tend to earn more than women earn in other
sectors of the workforce. According to a July 2009 survey, the
average starting salary for someone with a bachelor's degree in
mechanical engineering, for example, was just over $59,000. By
comparison, the average starting salary for an individual with a
bachelor's degree in economics was just under $50,000 (National
Association of Colleges and Employers, 2009).

:ath skills are considered essential to success in STEM fields.
Historically, boys have outperformed girls in math, but in the past
few decades the gender gap has narrowed, and today girls are
domg as well as boys in math on average (Hyde et al., 2008).
Girls are earning high school math and science credits at the same
rate as boys and are earning slightly higher grades in these classes
(U.S. Department of Education, National Center for Education
Statistics, 2007) (see Figures 4.1 and 4.2).
On high-stakes math tests, however, boys continue to outscore g1rls, albeit by a small margin. A small gender gap persists
on the mathematics section of the SAT and the ACT examinations (Halpern, Benbow, et al., 2007; AAUW, 2008). Fewer girls
than boys take advanced placement (AP) exams in STEM-related
subjects such as calculus, physics, computer science, and chemistry (see Figure 4.3), and girls who take STEM AP exams earn

. ES 2007 467) (Wash ngtor DC Govemment Pn. ra• or a ona Center for Educat on StatiSt cs 2007 The Nat1on s W!ool graduates ResulS from the 2005 NAEP Hyh School Tronsap.• g 0~1ce) Gr s • BOY' at FICiUI[ 4 2 Y r .Il 1 The ~ Is n the QuestiOn • G rls • Bo s 2000 1998 2005 Htgh School Graduateon Year FIGURE 4 l .

Students Takmg Advanced Placement Tests in Mathematics and Soence. 3.collegeboard. lower scores than boys earn on average (see Figure 4.. Source: Retrieved November 11.777 .D E :::> z 90. .686 I 08. .285 oL-----------------------------------~ Gtrls Boys Fl Gu RE 4.268 9. I • computer science AB • Physics c ..249 114. from the College Board website at www.. 2009.. .4 ). :. by Gender.. 4. Ql .CHAP T E R 4 Women and Girls in Science. • .< and magnetism • Computer science A • Physics C • Mechanics • Physics 8 • Calculus BC • Environmental science • Chemistry • Biology Calculus AB . Research on "stereotype threat" sheds light on the power of stereotypes to undermine girls' math test performance and may help explain the puzzle of girls' strong classroom performance and relatively weaker performance on high-stakes tests such as these. 2009...965 . .. 1 ~7 12.. and Mathematics 391. Technology. Engineering..867 64.

In 2005.0 . whtch neganvelv affects rh etr a 1 ttv ro enter . hm\e\er.. · d arc to have access to advanced less ltkely scian mat tn h · · b'J' ence m htgh sc ool.. H~{pern. Perna eta . from the College Board website ai \\WCO ~geboardcom One notable gam i~ girLs' mcrea cd repre entation in the ranks of the highest achiever in mathematic . et al..0 "' Ill 0u VI 2. ince the earlv 1980s the ratio of boys to girls in this extremeh sc. 4. 31 percent of Astan \ ' mencan and 16 percent . 1995).000 students). y co~plete STEM majors 111 college (~la}· & ChuI bm. boy continue to outnumber gtrls (Lubmsk.. . St 1 nlev 19 '3) ro around 3:1m recent years (Brody & \lJIIs. the proporrion of girls among the highe t math achievers has greatly increased during rhe pa t few decade . 2006: Hedges cr i Towell. Average Scores on Advanced Placement Tests mMathema•tcs and Soence Su Source.01 percent or 1 in I 0. . 2003.U N IT 1 The Answer Is in the QuJ~e~st~io~n~-----------------• GirlS • Boys 5. "'I 3. Among tudem~ with ven high o.5 ~ CJ to- ~ < 2. 2008..0 4. The Stud} of ~lathemancally Precocious Yomh identifies seventh and etghth graders who core greater than 700 on the AT math ecrion (rhe top 0. 200S.cores on m. Tyson et al 200..-..5 Ill ~ u VI . 2007)..uh le t. 1 d femal both can Amencan and Htspantc students • courses e an hma e...'' 2009). 1992. Benbow. 2009.5 4.lect group ha·-. Af · h Students from historically dtsad\anraged grou ps sue as n.· · .0 F1GuR E 4. dramancall> declined from 13:1 (Benbow &. f II and succe~s u.._ 3. Retneved November 11. Fnzell & Nave....t &.. Benbow..

many of these academically capable women leave STEM majors early in their college careers. or the physical sciences.5). National Center for Education Statistics. they are far less likely than their male peers to plan to major in a STEM field (see Figure 4. women leave STEM majors is still an important area Women make up a smaller number of STEM . 2009b). one-quarter of Asian American and one-tenth of white high school graduates took either the AP or International Baccalaureate exam in calculus. Just over one-fifth of male freshmen planned to major in engineering. compared with only about 5 percent of female freshmen (ibid. Xie & Shauman. Technology. a growing number of g1rls are leaving high school well prepared in math and science and capable of pursuing STEM majors in college. Department of Education.S.. Yet even among underrepresented racial-ethnic groups. Additionally. 2003). Nevertheless. Although the overall retention of female undergraduates in STEM is similar to the retention rate for men and has improved over time (U. 2008). National Center for tA. in engineering the national rate of retention from entry into the major to graduation is just under 60 percent for women and men (Ohland et al. Department of Education. 2000.S. compared with just 3. The gender disparity in plans to major is even more significant when the biological sciences are not included. compared with only 15 percent of all female freshmen. respectively. planned to major in a STEM field in 2006 (National Science Foundation.). 2007). and Mathematics of white high school graduates completed calculus. Female and male first-year STEM majors are equally likely to have taken and earned high grades in the prerequisite math and science classes in high school and to have confidence in their math and science abilities (Brainard & Carlin. compared with 6 percent and 7 percent of African American and Hispanic high school graduates.6 percent of Hispanic graduates (National Science Board. WOMEN IN STEM IN COLLEGES AND UNIVERSITIES The transition between high school and college is a critical moment when many young women turn away from a STEM career path. For example.UCilti4)i Statistics. as do many of their male peers (Seymour & Hewitt. 2008). computer science. 1998. Almost one-third of all male freshmen (29 percent}. Vogt et al. U. 1997)..CHAPTER 4 Women and Girls in Science. 2000. Women who enter STEM majors in college tend to be well qualified.2 percent of African American and 5. Although women are the majority of college students. Engineering.

..1% 0 10 20 30 40 Percentage .:...:::. D1vision of Science Resources Statistics.. Speoa/ tab t ( 50 Sctence Foundation..:.. 2009.....4% Female African American 16._~.-.3% • Mathematics/statistics sciences •• Computer Engineering 17....~~--~:-----'---..2% Female Asian Male 42..J ---------+-American Indian Female 29....:.4% Male 27.A~ ~----------------------------- All races/ ethnicities r • Biological/a-g~icultur:l Female Male sciences • Physical sciences ~----~~~=--. VA) Table 8·8..:.::::J 28...5% Female Male Female White Male 1D ~--... cited in National lsablltttes msoence and engmeering· 2009 (NSF 09·305) (Arlington... 5. Survey of the Amencan freshman. and Gender 2006 F1GuRE 4......nswer Is in the Question _ _ _u_N_I_T_l_T~h:.=e. .:..... minonties. Intent of First-Year College Students to Major 1n STEM Fields' by Race-Eth mctty · ' Source H1gher Educat1on Research Institute. 2007...gel~.30fo Male 29.... CA).... and perso~~ ~~ ~~os An...:.:.6% 6811 Hispanic Female of Mexican/ Chicano/Puerto Male Rican descent Other Hispanic 29. Women..:...1% 27.

o the lo~ of women from these ma jors i.trned the rn. 7 how . .ugolis. and nearly one-half 111 math..\l fields. As Figure 4. as Figure 4. of special concern.CHAPTER Women and Girls in Science.1!1e M. Engineering.uned a much smaller proportion of bachelor' degree . In fact. women earned 48. J. Technology. \X1omen e. and computer 1ence.M disciplines. varies dramatically.tjority of bachelor's degrees 111 biology. In the mid-1980s \Omen earned ~lightly more than one-third (36 percent) of the ha helor' degrees in computer science.tlthough women's repre-. and Allan Fi her. how the role of depurrmenrul culture in attracting and retaining female computer science and physics ITI<ljors.mnot he taken for granted. the num ber of degrees awarded. one-half of b. ''omen e.lfded in phy ics.wcally durmg the pa t four decades.entation 'ane b) field. and. • 1966 1976 • 1986 • 1996 • 2006 Computet saencc 2006 .6 shows. and Mathematics tart. therefore. De pite the still relatively small percentages of women majormg in orne STE.001 biological science degrees in women's representation in omputcr science is actually declining-a stark reminder that \Omen' progress c. engineering.uded to women h. The work of re earchers Barbara \X1hi ttcn. hy 2006 that number had dropped to 20 percent. the ovcr<tll proportion of STF###BOT_TEXT###gt;1 bachelor' degrees's degrees in chemistry.ts increased dr. The ize of the STE. In 2006.

944 34."' -. · · · Amencan/Alaskan Nanve N· · ncan Amencan ' Hispan1' c ' and at1ve) ge 1 nera ly mirror the overall .science degrees. ::J 6.000 48 001 31. 2.874 3.399 • • • • 120. bachelor's • Ywomen from underrepresentc d rac1a 1-cthmc groups (Af .. 7.371 90.109 dec~r~. z 60.~clenc~ and degrees. VA).652 computer science degrees 16 438 l c ectncal engineering degrees. In com· 0 og1cal ' . - - - - Physics earth. 34.017 0 2..846 physics degree. by Gender.678 2. physics degrees. atr:nospheric. . compared with only 7.000 7.347 Q L-----~----------~--~--------~ Women Men Fl GU RE 4.. Source Nat1onal Sc ence Founaat10r\ DIVISion of Soence Resources StatiStiCs. men earned 31 347 b'1 1• .'ll 5 8.827 . Cll Cll 88.614 Cll . and persons Wtth disabilities in sdence ' ' and engmeermg 2009 (NSF 09-305) (Arl1ngton. . .000 00 0 1.024 1.944 comp degrees.000 • • • • .109 5.743 2.enng Electrical eng1neenng Civil engineering Chemistry Mathematics and statistics computer sciences Agricultural sciences Biological sciences 1.652 B. trical engineering pari son. and 3. Bachelor's Degrees Earned 1n Selected Sc1ence and Eng1neenng Fields. and ocean sc1ences Chemical engin~ering Mechanical e~g1ne.U N I T 1 The Answer Is in the Question 150.781 30. Tables C-4 and C-5. 2007..499 Cll _ _ _ _ _ill.. ' degrees earned b Trends in.0 E . 2009 Women mmor·r· ltes.846 • I - 2.000 138.

physics.S.000 """"" ~ :s.CHAPTER 4 Women and Girls in Science. 2009. the overall number of Afncan American women earmng physical science bachelor's degrees was less than 600. atmospheric. and persons with disabilities m sdence and engmeermg: 2009 (NSF 09-305) (Arl ngton. pattern.9). minorities. 8. Forty years later. Data based on degree-granting institut1ons eligible to participate in Trtle IV federal financ1a1 aid programs. women earned almost one-half of the doctorates in the biological . :" "' w "' ""Do 2. DIVision of SCience Resources Stattstics. and Mathematics r. engineering. Bachelor's Degrees Earned by Underrepresented Raciai-Ethn1c Groups 1n Selected STEM Fields. cit1zens and permanent residents only. Engineering. and ocean sciences. however. Women.000 Ill 0 1. in some cases the gender gap in degrees earned by African Amencan and Hispanic women and men is much smaller or even reversed (see Figure 4. in 2006. and 3 percent or less of the doctorates in earth.n6 3. Women's representation among doctoral degree recipients in STEM fields also has improved in the last 40 years (see Figure 4. African American women earned 57 percent of physical science degrees awarded to Afncan Americans m 2007. Technology. Table C-14. women earned about one-eighth of the doctorates in the biological and agricultural sciences. still. 2007.8).000 Computer sciences Physical sciences Engineering FIGURE 4 . Source: Nat1onal Saence Foundation.::. 6 percent of the doctorates in chemistry and mathematics.000 -. For example. In 1966. by Gender. Note: Rac1al-eth'11C groups mdude U. l Women •Men 4. and computer science. VA).

VA). however.11~er any educatronal progra~ or actrvrty recetvrng federal frnancial assistance" (20 U. Ill D . be denied the benefits of. the . ecent efforts by Congress have brought attention to how Trtle IX could be used to rrrrprove the drmate for and representation of women rn STEM ftelds. federal agencres agenaes ndudrng NASA and the Department of Energy n conJunctron wrth the Department of Edu· t esponse to these frndrngs. Trtle IX can help create a chmate where women and men of srmilar talent who wa t t b g de~-specrfrc w_ays (Preronek. most Yto sports. or be subjected to drscr 111natron 1. n e soentrsts or engrneers have equal opportunrty to do so. on the ba~is of sex•. agricultural atmospheric. Code§ 1681 ). Author's analysts of Tables 34. Crrtrcs argue that women do not face drscnmrnatron rn STEM frelds but rather that women are less inte t d h . . 2oo8. ~e excluded from participation in. sciences and ocean soences Chemistry Mathematics Computer science Engineering Physics FIGU E 4. ed r R e uca ron at a eves. Table 25. ~oo6 _ 40 10 0 Biological and Earth. Trtle IX covers nearly all colleges a1d unrver5ttres. When Congress enacted Trtle IX. 1966 1976 • 1986 1 • 1996 • so I L. Science and engmeering degrees· 1966-2006 (Detatled Statrstrc. Tables) (NSF 08-321) (Arlington.w was ntended to help women achreve equal access to all aspects of d t. unng 1 toe last 37 years. A report by the u. f elds and that enforcement of Title IX could lead to a quota system mthe sciences (Tierney. To ensure compliance wtth the law. Source· "Jatlonal Sae11ce Foundatton DMston of Science Resources Statisttcs. affect perso~al chore~ rn een revrews c~n help rd~ntrfy 2005) Srmply put. Trtle IX has been app. 35. Title IX and Gender Equity in STEM ""••le IX of tre Educatron Arllendments of 1972 prohibrts sex discnmination in educatron programs and activities that receive federal financial assrstance.!ut onal poha~ and practrces that negatrvely. 38. & 39. 2008.9. Munro ~ ~ ~i ~n men 1 ~ certatn STEM nor proport :>nahty and rt cannot address gender gaps rn partrapatron due to personal chotces· however Till IX · ~ e IX requrres nerther quotas rnst. need to do more to ensure that col eges and unrversrtres receMng federal funds comply Wtth Trtle IX In r concluded that. and rn some cases rnadvertently.r.UN IT 1 The Answer Is in the ~Q~u~e~st~io~n~------------------- r. Doctorates Ear1ed by Women 1n Selected STEM Fields. Tttle IXregulations require institutions that receive any form of federal educatron fundrng to evaluate thetr current pohcres and practrces and adopt and publish grievance procedures and a policy against sex drscr mrnatron.S. Government Accountab. ty Offrce (2004) focused on Title IX tn STEM discipltnes and . The law states. have 2 9 ° .s. "No person rn the United States shall. 2009) ca ron and the Department of Justice. federal begun t~ conduct Tr e IX compi anee revrews more regularly (Pieronek.

African American women earned 2. The proportions were s1m1lar for Hispanic women and even smaller for Native Amencan women (National Science Foundanon. women's representation varies by discipline as well as tenure status. This is an impressive increase. engineering. women made up about 44 percent of the field. progress has been made. but women remain vastly outnumbered in many STEM fields. as Figure 4. and Mathematics and agricultural sciences. In the academic workforce.12A and 4.2 percent of the doctorates awarded in the biological sciences and less than 2 percent of those awarded in engineering. even in the biological sciences. women made up a mere 1 percent of engineers in 1960 and only about 11 percent of engineers by 2000 (see Figure 4. when women made up about 27 percent of biologists. Although \Omen have clearly made great progress m earnmg doctorates m STEM fields. m 2007. but women still make up only a small minority of working engineers. For example. 2009b). and ocean sc1ences. yet. and math. Among workers who hold doctorates. in 2000. chemistry. In fields such as the biological sc1ences. Engineering. computer sc1ences. especially engineering and physics. Technology. the physical sciences. women are snll underrepresented m many STEM professions. and approximately one-fifth of the doctorates in computer science. and physics. women's representation in . women have had a sizeable presence as far back as 1960. at the doctoral level women remmn underrepresented m eveq STEM field except biology. atmospheric. and mathematics and statistics.CHAP T E R 4 Women and Girls in Science. Overall. women's representation in the STEM workforce has also Improved sigmflcantly in recent decades. Forty years later. around one-third of the doctorates in earth. In general the number of doctoral degrees m STEM disciplines earned by women from underrepresented racial-ethnic backgrounds also increased during the past four decades but snll remains a small proportion of the total. Forty percent of the full-time faculty in degree-granting colleges and universities in the United States in 2005 were women. men represent a clear majority in all STEM fields. however.12B show that men far outnumber women.11 ).10 shows. Figures 4. On the other end of the spectrum. WOMEN IN THE STEM WORKFORCE Consistent with the mcreased representation of women among STEM degree rec1p1ents.

ltlon 1 even more severe for \Vomcn from underrt:prc ntcd racial-ethnic background . E'en m the haologacal sciences the number of Afr•· an Ameracan and Ha pamc female faculty was low.150 postsecondary teache~s an enganecrmg. 2008). DQ. Women made up le s than one-quarter of the faculty in computer and mformation ·1ence (22 percent). an area m which many people assume that women have achieved p3fll).\t d1sciplincs was significantly lower. In all ca e women were hctter represented in lower faculty ranks rhan 111 h1gher r.. onlv 60 were African men an \omen. TE.aence doctoral faculty in 2006. the ph) steal sciences ( 18 percent). \Omen made up only one-third (34 percent) of the faculty. In the life sciences. Of the more than 7. Of the nearly .U N I T 1 The Answer Is in the Question 60 B 'Labo· Sta• 51 cs 2009.:ompurer. The uu. African American women also mad~ up lc th~n I percent ?f rh~ 17.tculty in four-year colleges • nd urmer me (D1 htbio et al. math ( 19 percent).ere too low to report. Women m the labor force A databook (Report 1018) (Washmgton..000 \.. numl cr for Hispanic and· Native American ''omen . and engineering ( 12 percent).111ks among TE~1 f.

. . so postsecondary teachers are not Included here... & 2000. in most STEM fields the drop-off is pronounced. When postsecondary teachers were mcluded from 1960 to 1990.9% !! c: IU ~ 10 IU a.... ~ • Mathematical and computer scientists' • Chemists• • Physicists and astronomers' • Engineers I ___ j IU 0 i. the general trends remained the same. ' In the 1980 and 1990 censuses.. especially at the tenured faculty level m science and engineering. and gender disparities in representation will disappear.~ E< wo -. If we compare the percentage of tenured female faculty in 2006 with the percentage of STEM doctorates awarded to women in 1996 (allowing 10 years for an individual to start an academic job and earn tenure). and secondary school increases (more girls go into the pipeline).. no category for computer scientists was included. in the 1970 census.~ - IU 0.S. 1n the 2000 census. the category was titled "chemists and matenal soertJsts.. data 1nclude life scient1sts as well as biological scientists. 25. For example. 1970.s:: 20 ~~ tlO 13..C HAP T E R 4 Women and Girls in Science. women .000 postsecondary teachers in the biological sciences. "''"". 1960 1970 1980 1990 2000 Fl GU RE 4. Engineering. In the 1960 census.. the category was titled "physicists. 1960. 1990. The path from elementary school to a STEM career has often been compared to a pipeline." 1 Source: U. 380 were African American women and 300 were Hispanic women (ibid. the category was titled "chem1sts except biochemists". DC).. .. 353% c: IU ~e / r.11.. 1960-2000.1% "'c: . .~ "'"' 41 7% 40 ~ 0 ~ Q." 1 In the 1980 and 1990 censuses. 1980. Census Bureau.. middle. 2002). Women's representation among tenured faculty is lower than one would expect based on the supply of female science and engineering doctoral degree recipients in recent decades (Kulis et al. and Mathematics 50 ~ 44. the category was titled "mathematicians and computer speCialists. Note Data on postsecondary teachers by field of instruction were not gathered 1n the 2000 census. This has not happened at the expected rate. Technology. This metaphor suggests that as the number of girls who study STEM subjects in elementary. Census of the population (Washington.. Women 1n Selected STEM Occupations. the number of women who become scientists and engineers will also increase (more women come out of the pipeline).)." 4 In the 1960 census.

and Environm Gender and Employment Status.on of Sc1ence Resources Statistics 2009 Charoct . Source Nauonal Soence Foundation. Agricultural. Workers with Doctorates in the Computer and Information Sc1ences Workforce. Auth~rs' anal ensttfcs of doctoral scientists and engineers m the ' ys1s o Table 2.U N 1T 1 The Answer Is in the Quest~io~n~------------------ Male part time _ ____. by Note The percentages do not equal 100 due to roundmg. 3% Female full time 15% Female part llme 20/o Fl GU R E 4. Male unemployed lo. Authors' analysis of Table 2. Source Nat1onal Science Foundation. Workers with Doctorates in the Biological. Div1sion of Science Resources Statistics. by Gender and Employment Status. 2006. . Note· T1e number of female uneMployed workers was not available due to small sample size. Umted States· 2006 (Deta· ed Stat1st1cal Tables) (NSF 09·317) (Arlington. 2006."ables) (NSF 09-317) (Arlington. enta Life SCience Workforce. VA). VA). 2009.b Female full time 29% Female part time 4% Female unemployed 1% F1GuRE 4.12 B.11A. OIVIs. 2006 (De:ailed StatiStiCal". . 1 . Characteristics of doctoral scienttsts and engmeers in the Untied States.

Nelson & Rogers.13. where women now receive about onehalf of doctorates and received 42 percent in 1996. Technology.C HAP T E R 4 Women and Girls in Science. DIVIS on of Scsence Resources Stat1sUcs. Sotlct ona Soence Foundation. .13). n.Tenured faculty Nontenured faculty Engineering Phys1cal sciences Computer and anformation sc1ences B•olog1cal. nontenured STE. and environmental l1fe sc1ences 418% 0 10 20 30 40 50 Percentage of Faculty Who Are Women fl c. \Vomen make up larger percentages of the lower-paying. 1998. Kulis et al. 2003. smaller percentages of qualified women apply for these positions in the first place (National Research Council. Authors' analysts of Table 20. Although recent research found that when women do apply for STEM faculty pos1t10ns at major research universities they are more likely than men to be htred. 2002). 2006). Several studies have found a gender difference in hiring in STEM acadcm1c disciplines (Bentley & Adamson.. 2006. 2009. VA). by Disopline and Tenure Status. Improving women's position among STEM faculty will apparently require more than simply increasing the pool of female TEM degree holders (Valian. Female STEM Faculty m four·Year Educational Institutions.M faculty positions (see Figure 4 . ~ - . 2008. Cathy Trower and her colleagues at the Collaborative on Academic Careers in Higher Education (COACHE) at Harvard University found that female STEM faculty express lower job satisfaction than do their male peers. agriCultural.. Lower satisfaction leads to higher turnover and a loss of talent in science and engineering. 2009). uIE 4. and Mathematics earned 12 percent of the doctorates in engineering in 1996 but were only 7 percent of the tenured faculty in engineering in 2006. 2009a). Ginther & Kahn. Characteristics of doctoralsoentists and engineers in the Unltd Stctts 1006 (Deta ed Stausucal Tabes) (NSf 09-317) (Arhngton.d. Engineering. Even in fields like biology. women made up less than one-quarter of tenured faculty and only 34 percent of tenure-track faculty in 2006 (National Science Foundation.

nd hence the retention of ' . Women workmg m . quit their jobs by midcareer (about 10 years into their careers). The studies highlight midcareer as a critical time for these women. desptte similar levels of stated satisfaction and edu:atton. f . surveyed were eithei· not emp loyed at a ll or not .Improve JO b sans acnon a ' cies can fields tend to have higher earnings female STE~1 fac~lty. and technologists m bustness and the high-tech industry have found that women in these ftelds have higher attrition rates than do both their male peers and ###BOT_TEXT###gt;vomen in other occupations (Hewlett et al.. c:. In engmcering. help . many STEM careers can provt?e women .1 ' • • WHY SO FEW? Academic rc carch on th 1·5 to .. f . women have higher attrition rates than thetr male peers have. engineers. w h'l 1 eon y one-tent o h d d men surveyc a left the engineenng field. PIC IS I Ic. . 1 " occu . .sTEM workforce.mate of science and engihe c l1 • t that o f fema Ie t d to satisfaction Trower's research suggests 1 I e po1·tneering departments ·IS cIosel} ... compared with only 17 percent of their male em~loyees. 0 ne-quarter of an engmcenng · fema Ie engmeers . gap extsts 111 1 for in 2009 the <nerage startmg sa a$ry OOO a year and bachelor's 42 ' k t' g w·'1 s JUSt over · d t ting ' salaries averag. The Soctcty of Women Engineers (2006) conducted study of more than 6. . I d h f l emp oye m engmeermg or a related field .U N IT 1 The Answer Is in the Qu:es~t~io~n------------------ . As these numbers mdicate. High-tech companies in particular lost 41 percent of their female employees. and technologists are fairly well represented at the lower rungs on corporate ladders (41 percent). Recent studies of scientists. ye·u In comparison. For example than do other women 10 the patwns as m bachelor's degree recipi-' STE'1 . 2009). and average starting salanes wer~ JUSt u~der $6~.helor's degree holders in compu~er scten~e average aroun $61. engineers. Hewlett et al.rea e ntoring and wor k-]'f · ffecnve me 'd' facultv and that prO\t mg e . mg aroun bac.reased earning potential and greater economic secur. however. 000 1'nd'1vt·dua 1s w ho earned a retentton • • degree between 1985 and 2003 . 2008. degree recipients tn accountmg recetve s ar d $48 500 . startmg sad1anes ord . . Simard et al.Ity. 2008). · • ' ll'St t ' ' lc notion that men are math· .. with three themes I pro l'f' emerging from the litcr·lttirc 1. mc. More than half (52 percent).. ents m mar e tn .500. (2008) at the Center for Work-Life Policy at Hanard University found that female scientists.. although a gender pay · other fields.000 for mdt\ iduals holding bachelor's degrees m chemical engmeering (Nattonal Association of Colleges and Employers.

and J-Stor.C HAP T E R 4 Women and Girls in Science. with issues ranging from work-life balance to bias. Halpern. Soc1al Soence utatJon Index. and Mathematics ematically superior and innately better suited to STEM fields than women are remains a common belief. ProQuest. These proJects were chosen because they each add'ess an 1mportant rssue the potent~al to tnfluence public understand ng. although the connection between spatial abilities and success in STEM careers is not definitive (Ceci et al.. research shows that spatial skills can be improved fairly easily with training (Baenninger & Newcombe. as well as some tests involving memory and perceptual speed (Hedges & Nowell. 2002. remains hotly contested. that is. 2008). ~nclud1ng Web of Soence.. a difference in average math performance between girls and boys no longer exists in the general school population (Hyde et al.o&uca•~. that girls and boys tend to have different cognitive strengths and weaknesses. 1995. with a large number of articles addressing cognitive gender differences as an explanation for the small numbers of women in STEM. number of Citations. however. especially writing. soc•ology. Nevertheless.. MUW reviewed hundreds of academ1c arucles wr1tten dunng the past 25 years on the topic of women msc1ence and engineenng. Whether or no t welldeveloped spatial skills are necessary for success in science and engineering. Poor or . Voyer et al. Cognitive Sex Differences Methodology Usmg nultJp!e databases. Vasta et a!. A second theme revolves around girls' lack of interest in STEM. et al. neurosoence. education. Kimura. with boys consistently outscoring girls (Linn & Petersen. The I terature rev~ew 1nformed th s chapter. " 2 Other researchers have found. 2007). Technology. 1985. The prof1!ed f1ndmgs are well respected 1n the research communrty. Sorby and Baartmans (2000) and their colleagues designed and implemented a successful course to improve the spatialvisualization skills of fi rst-year engineering students who had poorl >· developed spati al skills. ArtiCles from the fields of psychology. neither girls nor boys are the "smarter sex. Generally. Girls outperform boys on tests relying on verbal skills.. economics. students who did not take the course. 1995). A third theme involves the STEM workplace. and oCher forms of public recognrt1on. and endocrinology were exam ned.. Aronson. 1996). compared with about one-half of the &\. Many people consider spatial skills to be important for success in fields like engineering. boys perform better tasks using spatial orientation and visualization and on certain quantitative tasks that rely on those skills. the issue of cognitive sex differences.. Engineering. Lynn and Irwing (2004) found small or no differences in average IQ between the sexes. The remainder of this chapter summarizes and examines these themes. including mathematical ability. Among the most promising research findings in this field are those of Sheryl Sorby. One of the largest gender gaps in cognitive skills is seen in the area of spatial skills and specifically on measures of mental rotation. As noted earlier. as measured by pubhca!lon rn peer· r~ Journals. These PfOit!dS were conducted wrthrn the pell15years. More than three-quarters of female engineering students who rook the course remained in the school of engineering. 1989.. 2009).

d'ffercnccs in bratn structute and hormones IS 1esea1c on sex 1 • d d h · 11 d' inconclusive. This logic has two mai~~~wer ~ women in c~rtain STEM gtrls have made rapid inroads · t hs." such . . That r 3? . found that hormonal exposu re. ~s tca Y. Ceci et al. the researchers concluded. computer science. Scientists and Engineers Are Not N . far faster than 1t would . 2008).ro s~ec1ftc . . the Highest Math Achievers ecessar•1y Boys outnumber girls at the very h' h dtstnbution. Dtfferences between countr1·es an d over ume . Differences in the representation of women in science and math fields cross-culturally and over time also support the role of sociocultural factors for explaining gender gaps in these fields (Andreescu et al. cogmttve strengths and weaknesses remains unclear. ev1~enc~ for cognitive sex differences based on hormon al exposu:e ts m1~ed. (2009) reviewed 11101 e · · STEM fields includcauses of women's underreprcscntatiOll 111 d I I d 'th I • · 1 f crors an cone uc e at t 1c ing biologtcal as well as soc1a a . hematl· 'll Importance of culture in the de ve1opment o f mat ca1 sk1 s. . 1 crate the . whde m the vast maJonty of countries more boys than girls scored above the 99th percentile in mathematics on the 2003 Program for In~ernational Student Assessment. bur how these differences translate in. does have a role in cognitive sex differences. Also. than 400 arttclcs explormg the Cec1 ct al. Ftrst. · . . 224). while women's ICS. ' . . the ratio of boys to girls among children identified as mathematically precocious ~as decreased dramatically in the last 30 years.Istinct.ta~e a genetic ch. . and en C~landtng ftelds SUCh as physg ncenng has grown slowly. Ltkewtse. suing math or science courses 1 spatial skills may deter girls fl 01~ ~ur roved fairly easily.. as ment10ned above. but these skills can e p Biology Is Not Destiny . Female and male brains arc Ill ee P. As discussed earlier. espectally 10 gestation. 1111 or gender differences in preferences and sociocultural influences on girls' performance on gatekeeper tests ( to travel through the population. in Iceland and Thailand more gtrls ~han boys scored above the 99th percentile (Guiso · 'II uset a!. "Evidence for a hormonal basis of the dearth of female scientists" is "weaker than the evidence for other factors. 2008). Overall. however.- U N IT 1 The Answer Is in the Question . Some researche h tg end of the math test score rs avebsuggested t hat t h'1s gender d'f 1ference accounts for the small 0 ftelds.h . as "mathematically gi fted'' inmho t e ranks of children identified ~epresentation in mathematical~ edpast ye~rs.

life sciences. Technology. and physical science workforce scored higher than 650 on the SAT math exam. Even though a correlation exists between high school math test scores and later entry into STEM education and careers.. The work of Shelley Correll shows that girls assess their mathematical ability lower than do boys with equivalent past mathematical achievement. Because of this. 24 percent of boys but only 5 percent of girls said they were interested in an engineering career. Second. and social sciences than in math. where boys are considered to excel. 2004. computer science. very high math scores are not necessarily a prerequisite for success in STEM fields. In studies of high mathematics achievers. Weinberger (2005) found that the science and engineering workforce is not populated primarily by the highest-scoring math students. Correll. women are more likely to secure degrees in the humanities. From early adolescence. fewer women pursue STEM careers than would be expected based on the number of girls who earn very high math scores.. 2008). Another recent poll found that 74 percent of college. "Just Not Interested" Many girls and women report that they are not interested in science and engineering. Interest in an occupation is influenced by many factors. 2006).. therefore. Engineering.. with girls reporting less confidence than . and more than one-third had SAT math scores below 550the math score of the average humanities major. computer science. girls are less likely to believe that they will succeed in a STEM field and. Eccles. Turner et al. including a belief that one can succeed in that occupation (Eccles [Parsons] et al.CHAPTER 4 Women and Girls in Science. 2009). girls hold themselves to a higher standard in subjects like math. math. engineering. 2000. bound boys ages 13-17 said that computer science or computing would be a good college major for them compared with 32 percent of their female peers (WGBH Education Foundation & Association for Computing Machinery. 1983.... Pajares (2005) found that gender differences in self-confidence in STEM subjects begin in middle school and increase in high school and college. male or female._ _.. Less than one-third of college-educated white men in the engineering. and Mathematics is. In a 2009 poll of young people ages 8-17 by the American Society for Quality. or the physical sciences. Even girls and women who excel in mathematics often do not pursue STEM fields. girls express less interest in math or science careers than boys do (Lapan et al. At the same time. the reverse is true for men (Lubinski & Benbow. are less likely to express interest in a STEM career. 2006) . for example.

source of t h ts that the difference is innate wht'le uh )ect o . A review of child vocational development by I lartung et a!. A belief that one can succeed in a STEM field is important but is not the only factor in establishing interest in a STEM career. Students who lack co~fidence '· ' 1or sctence ' · math ·111 rhetr .. The g ts e a . Sax 199 4 ) ~ va ~e makmg a social contribu· a clearer social purpos~ such · be. Girls and women may be especially vulnerable to losing confidence in STEM areas. . e ability In part.U N I T 1 The Answer Is in the Question . engage sk'll rhat require those skills and will more quickly . ebate: Some claim · a resu It · IS ot ers clatm t hat It of gender socialization · Reg'ar dl ess of · · f th ~ ongm o the difference. Mar_ . As a ~:sul~ademy of Engineering.. 1994. 2005)..g1ve up m the face of difficulty. 2005). 2006). boys develop boys do in their math and scienc h exp. .s k1 s... 1994. I ' .careers often do not appea l to women (or men) wh tion (Eccles.ll A hen variables such as previous evant .. . a leading researcher in the field of occupational ~h01ce~ has spent the past 30 years developing a model and collectmg evtdence about career choice. Eccles. In tasks 1 s are Jess likely to. Ecc1es 2006). on doing work that contn· bu tes to soctety. ' 1986· Z1mmerman & Martmez-Pons. ' fd gender d'tfference is a s b. . n~m er . Culturally prescribed gender roles also influence occupational interest (Low et a!. Pa'ares 1996. · h and men place .to. Her work suggests that occupational choice is influenced by a person's values as well as expectancy for success (Eccles [Parsons] et al. son '199 1. have succeed I~a engmeering and env•· ed .. (2005) found that children-and girls especiailydevelop beliefs that they cannot pursue particular occupations because they perceive them as inappropriate for their gender. 1983.rram STEM subdisciplines with . Jacqu~lynne Eccles. 2000 . w cnccs tn self-confidence dtsappear learn are contra · 1990· Cooper & Robinachicven:ent or opportun ity . .attracting higher per· centages of women than have h or electrical engineering (Gibbot er subdtscrplines like mechanical ons. ronmental engmeering. h a c1ear men to p refer work wtt wtt· 1women more likely . than d Konr 1993· al. et soc!a purpose (Jozefowtcz a et a 1.. as tomed. most people do not view STEM society or individuals (Natio o~cupatlOns as directly benefiting 1 Diekman et al. u bms I & Benbow' 2006 . The ~esearch of Car:ol Dweck has implications for improving self-conf1dence. . I . 2009). Well-documented gender differences exist in the value that · women . k' J t 2002 01 L . I lied (Lent et a. STEM . 2009). 2008.m . Dweck s research shows that when a girl believes that she can become smarter and learn what she needs to know in STEM subjects-as opposed to believing that a person is either born with science and math ability or not-she is more likely to succeed in a STEM field.erience developing reigreater confidence in STEM dt?rouhg shown that gender differb of stu tes ave .

can lead to success in STEM fields (Hanson. The number of African American women in STEM remains low.engi neeryourlife. 2005). . Another ongoing study and outreach project is focusing on educating high-achieving. unusual career for women while emphasizing the people-oriented and socially beneficial aspects of engineering. many of the characteristics that are considered appropriate for African American women. Although the girls knew almost nothing about engineering at the start of the study. (2009) reported an increase in middle school girls' interest in engineering after the girls were exposed to a 20-minute narrative delivered by a computer-genera ted female agent describing the lives of female engmeers and the benefits of engineering careers. race. and assertiveness. Of course.C HAP T E R 4 Women and Girls in Science. 2004). Engineering. and ethnicity simultaneously. for example. 2005. gender and race do interact to create different cultural roles and expectations for women (and men) from different racial-ethnic backgrounds. In the African American community. high school girls about what scientists and engineers actually do and how they contribute to society. has also been shown to increase high school girls' interest in pursuing engineering as a career. of the 66 percent of girls still participating after two years. 2008. 2009). mostly minority. suggesting that other barriers are important for this community (ibid. Fouad & Walker. 2004. Young African American women express more interest in STEM fields than do young white women (Hanson. 2008 ). Although these studies generally relied on small samples and in a number of cases no longterm follow-up has been done with participants. the results are promising. Assumptions about the mismatch between women's interests and STEM often are based on the experiences of white women. In a survey by Paulsen and Bransfield (2009). Technology. Plant et al. independence. The Engineer Your Life website (www. and 76 percent said that it inspired them to take an engineering course in college. such as high self-esteem. however. recent research suggests that there are ways to increase girls' interest in STEM areas (Turner & Lapan. 80 percent were seriously considering a career in engineering (Eisenhart. 88 percent of 631 girls said that the website made them more interested in engineering as a career..). a project of the WGBH Educational Foundation and the National Academy of Engineering. Eisenhart. and Mathematics Despite girls' lower stated interest in science and engineering compared with boys. Research on interest in science and engineering does not usually consider gender. The narrative included positive statements about students' abilities to meet the demands of engineering careers and counteracted stereotypes of engineering as an antisocial. Plant et a!.com ).

wo ' · ' k 1 than do their male peers (Society . bias. (2008) found that many women appear to enc~unter a series of challenges at midcareer that contribute to their leaving careers in STEM industries. 1 . Wor p ace environ]] la a role. Nosek et al. Xu (2008) showed that female and male faculty leave at similar rates.der and. 2008. Goulden et al.h ~) found that majorities of both association of male with sc· et me groups hold a strong implicit . Bias. Women cited feelings of isolation. . advancement opportunities. . tas that negatively mfluences their progress a d explicit bias may be dec~e ~artt~tpa~lOn.. rna e Wtth liberal arts. As mennoned above. .. 1998). . percent less likely to enter a tenure-track positiOn after recetvmg a doctorate. (2009) compared men and women in the sctences who are married with children and found that the ~omen were ~5. Bias . however. tases may reflect b . Women in STEM fields can experience b. Therefore. an unsupportive work environment. tence and fe l R esearch has also pointed t0 b" . ' e stronger than. . For example Wen empreJs et al. the situation in academia is somewhat more nuanced. In a recent study on attrition among STEM faculty. tmpbcit bias continues to have an . 1999· Trix & Psenka. neras and Wold fou~d that a female ' 2 . Women's higher turnover intention in academia (which is the best predictor of actual turnover) is mainly due to dissatisfaction with departmental culture. faculty leadership.. Implicit b"asmg. Although women and men in industry and business leave STEM careers at significantly different rates. ' ment. 1997) and hiring (St · 2003). and gtrls in science and math women and men of all racial. and research suppo~t. extreme work schedules. ( ~~. and Family Responsibilities . . of Women Engineers 2006· . or lfl some cases contradict explicit! h ld even individuals who espouse y b ~· f beliefs or values.. hence. 2009). and family responsibthnes a P Y Workplace Environment In the study of STEM professionals in the private sector described earlier Hewlett et al. Although instances of adverse effect. . as m peer review (Wenneras & Wold. Hewlett et aI. men leave STEM fields at a htgher rate . ity may harbor implicit biase a be te of gender equity and equalgender stereotypes about wos a out gen.iiL_~U~N~IT~l~~T~he~A~n~s~ Workplace Environment. and unclear rules about advancement and success as major factors in their decision to leave. negative (Valian.w er~l~s~in~th~e~Q~ue:s~ti~o~n------------------------------------.. Frehill et al. women are more likely than men to consider changing jobs within academia.

even when these women have proved themselves to be successful and demonstrated their competence (Heilman et al. while the physical or hard sciences and engineering fields are still considered masculine domains (Farenga & Joyce.. While nothing is wrong with being compassionate. in choosing what features to include in their profiles of the female applicants" (p. This meant that she either had to publish at least three more papers in a prestigious science journal or an additional 20 papers in lesserknown specialty journals to be judged as productive as a male applicant. teaching. The authors concluded. and ability for success in acadc:uJLl·. Engineering. trying hard. and being a good teacher. medicine. "Recommenders unknowingly used selective categorization and perception. Letters written for women are more likely to refer to their compassion. These results suggest that gender stereotypes can prompt bias in evaluative judgments of women in male-dominated environments. 1999). women are less well liked and more personally derogated than are equivalently successful men. 2004 ). and ability. 2009. Technology. leading to lower evaluations and less access to organizational rewards. research. which are the characteristics highlighted for male applicants. for example. The authors concluded that the systematic underrating of female applicants could help explain the lower success rate of female scientists in achieving high academic rank compared with their male counterparts. 215). and effort as opposed to their achievements. and Mathematics postdoctoral applicant had to be significantly more productive than a male applicant to receive the same peer review score.CHAPTER 4 Women and Girls in Science. Biases do change. When women are acknowledged as successful in arenas that are considered male in character. women are not expected to have significant accomplishments in a field like academic medicine. Life and health sciences are seen as more appropriate for women. The researchers concluded that recommenders (the majority of whom were men) rely on accepted gender schema in which. also known as stereotyping._____. Being disliked can affect career outcomes. Trix and Psenka (2003) found systematic differences in letters of recommendation for academic faculty positions for female and male applicants. Xie & Shauman. research. however. arguably these traits are less valued than achievements. 2003). research . Family Responsibilities Many people think that women leave STEM academic careers because they cannot balance work and family responsibilities (Mason et a!.. Today the fields viewed as stereotypically male have narrowed considerably compared with even 30 years ago.

Married women Ill STEM appear to ha~e a disadvantage ~ompared with married men in relation ro tenure and promotion decisions only if the married women have c. w' cry o f women Engmeers 2006 .. both women an d men 1 entt y am1 1y ·b·1· · and industry respons1 1 1t1es'ffas a poss 1ble barrier to ad vancement b ut women d ' . · 1 · · b · Research shows that etng smg e IS u \-til be hired for a tenure-track job and promoted. that marriage is a good predictor f?r both women and men of being hired as an assistant professor (~1e & Shauman. . and not always worse in.. having young children does affect their chances for advancement. . I d 'ld h o mg marnage r c _1 ren an de aymg d having children. Research also shows. STEM f1e~ds. ln business . In another Mason and Goulden study (2004). 2003· Ginther & Kahn 2006). The same study found that among science professors who had babies wnhm the first five years after receiving a doctorate. .U N IT 1 The Answer Is in th e Question evidence b Xu (2008) pomts to a more nu~nced relationship STEM careers · ·b'l · . . So while marriage does not appear to hurt women.. are a ffected d1 erently than men by this "f . Additionally. F h. d academJc > h .1 ami Y penalty (Stmar hl h et al 2008 p 5) Alth f oug )Ot women d . an men eel that havmg a famdy hmders thet· l'k r success at work ' women are more 1 e1y . . ' 11 y-re a ted Issues (Soclwomcn in STEM are m~re lik '1 re ~II et al. 2003). where a " two body probl man . wo r k sc hed ule. .htldren (X1e & Shauman. 77 percent of the men but on ly 53 percent of the women had a~hiev~d tenure 12 to 14 years after earning a doctorate. women ~ arc more likely to report that th a_re the pnmary caregiver and have a partner who also works found th at most women and u tlhme. than men to report forego· . 70 percent of the men but only 50 percent of the women had children living in their home. Having young children in the home rna} affect women's productivity since child-care responsibilities fall dtsproportionarely on women (Stack. 2004 ). STEM and faces a similarly dee y tod ' ave a partner who is also in . In a situa. another career was amen w o left engmeenng . however. These ~1spannes were nor unique to.d 'f f . more than tw1ce as many fe~ale academics (38 percent} as male academics (18 percent) md1cated that they had fewer children than they had wanted. sa1'd t hat . mg tJon em ex1sts h · · . Some telling statistics point to the difficulties that mothers still face in an academic environment. 't e man's career is often g1ven pnonty (Hewlett et al.. ' : . l fan 'I an tme . am1 tes. A recent retention study .. Among wo men an men w th f . between farmh respons1 1 ttJes an at a woman t ctor pred1 good . . Mason and Goulden (2002) found that among tenured faculty in the sciences 12 to 14 years after earning a doctorate. 2008).mtercst m l1"keIy than men to also eire t' reason d ' bu t women were far more . /iJ . 200S)..

Chapter 2 profiles research showing that when teachers and parents tell girls that their intelligence can expand with experience and learning. girls hold themselves to a higher standard than boys do in subjects like math. multiple solutions are needed to correct the imbalance. and mathematics. and Mathematics WHERE DO WE GO FROM HERE? Multiple factors contribute to the underrepresentat ion of women and girls in STEM and. Research shows that negative stereotypes about girls' abilities in math are still relevant today and can lower girls' test performance and aspirations for science and engineering careers. Technology. as an obstacle to women's success in STEM fields. Girls' Achievements and Interest in Math and Science Are Shaped by the Environment around Them This report profiles four research projects that demonstrate the effects of societal beliefs and the learning environment on girls' achievements and interest in science and math. the difference in performance disappears. Selected for their relevance to public debate and their scientific credibility. engineering.C HAP T E R 4 Women and Girls in Science. At the same time. girls do better on math tests and are more likely to want to continue to study math. The findings are organized into three areas: social and environmental factors that shape girls' achievements and interest in math and science. Engineering. demonstrating that social and environmental factors clearly contribute to the underrepresentat ion of women in science and engineering. therefore. often operating at an unconscious level. One result of girls' lower self-assessment of their math abilityeven in the face of good grades and test scores-and their higher . the college environment. The remainder of this report profiles eight research findings. each of which offers practical ideas for helping girls and women reach their potential in science. When test administrators tell students that girls and boys are equally capable in math. technology. and the continuing importance of bias. Research on self-assessment finds that girls assess their mathematical abilities lower than do boys with similar past mathematical achievements. illustrating the importance of the learning environment for encouraging girls' achievement and interest in math. these case studies provide important insights into the question of why so few women study and work in many STEM fields. believing that they have to be exceptional to succeed in "male" fields. These findings provide evidence on the nurture side of the nature-nurture debate.

College and university administrators can recrmt and reta1~ more women by implementing mentoring programs and effective work-life policies for all faculty members. Implicit bias may · fl d f . ·d h . gf atial skills. g consistent and men consistently outperformmg . Research provides evidence that women are less satisfied with the academic workplace and more likely to leave it earlier in their careers than their male .U N IT 1 The Answer Is in the Question f wer irls than boys aspire to . g standard for performance 1s that e d lar est. 111 an environment that en ances t 1e1 they are more hkely to develop the1r · · · 1 k'll ·1 STEM · f . sim a w1t nme short a in matically 1 ·r success m science an math h .harvard. n ~mong peop e w o support gender equity. 10 1 sk1lls as well as their confidence and cons1der a uture a field. Bias.area ~lss~nd women. Takin h . 1 1 ··Is are . an 1 cognitive abilities is found n the . many girls graduate from high school well prepared to pursue a STEM career.counterpa~ts are. .edu can hel g t e m~hc1t b1as test at https:/1 their own implicit biases sot: p~ople Identify and understand at t ey can work to compensate for them. can add up to big gains in female student recruitment and retention. Often Unconscious.d . but few of them major in science or engineering in college. . m uence gtrls' likel'h 1 oo o 1 ent. and providing a student lounge. k'lls . . 1 implicit. colleges and universities can attract more female science and engineering faculty if they improve the integration of female faculty into the departmental culture.t ymg with and participating in math and sc· Ience and also contnbutes k to bias in education and the l h wor p1ace-eve .f . . Little changes can Make a Big Difference in Attracting and Retaining Women in STEM As described earlier. Research dra· 1y 1mprove . STe M careers. Limits Women's Progress in Scientific and Engineering Fields Research shows that most peopl 1· and math fields with "male" and e cont ~~e to associate science human!nes and arts fields with " female " includin · d. course. such as changing admissions requirements. g m lVI ua1s who actlv I ' e Y reJect t ese stereo. presenting a broader overview of the field in introductory courses. At Colleges and Universities. Likewise. types. . with boys One of the most consls~ent. w1t1 spatia s 1 s rrammg. Small improvements in the culture of computer science and physics departments. gender differences in . If gtr documents that individuals h · spie trammg · spatia · d · .

Defined by occupation. 2005). Those in science and engineering occupations who had bachelor's degrees were estimated at between 4.CHAP T E R 4 Women and Girls in Science. 2010). 2. The following eight research findings. Estimates of the size of the scientific.. like scientists or engineers. . government agencies including the Census Bureau. engineering. women in STEM fields can find themselves in a double bind. the National Science Foundation. the size of the STEM workforce has been estimated to exceed 21 million people.3 and 5.3 and 5. and the Bureau of Labor Statistics. Some research suggests that women and men achieve similar IQ results using different parts of the brain (Haier et al. Defined more broadly. and Mathematics People not only associate math and science with "male" but also often hold negative opinions of women in "masculine" positions. the United States science and engineering workforce totaled between 4. The National Science Foundation includes social scientists but not medical professionals in these estimates (National Science Board. Because both likeability and competence are needed for success in the workplace. Engineering. suggest that creating environments that support girls' and women's achievements and interest in science and engineering will encourage more girls and women to pursue careers in these vital fields. Women have made impressive gains in science and engineering but are still a distinct minority in many science and engineering fields. taken together.0 million. she is considered to be less likable.S. When a woman is clearly competent in a "masculine" job. Notes 1.8 million people in 2006. Technology. and technological workforce are produced using different criteria by several U. This research shows that people judge women to be less competent than men in "male" jobs unless women are clearly successful in their work.