Delusions of Gender (38 page)

1
(Brizendine, 2007), p. 161.

2
A claim made in the blurb of Brizendine’s book.

3
(Baron-Cohen, 2003), p. 2.

4
The Autism Research Centre was the source of the Empathy Quotient and Systemizing Quotient questionnaires:
http://www.autismresearchcentre.com/tests/default.asp
.

5
See (Baron-Cohen, Knickmeyer, & Belmonte, 2005), table 1, p. 821. Sixty percent of men report an S-type brain, compared with 17 percent of women. (Percentages include ‘extreme’ E and S brain types.)

6
(Schaffer, 2008), entry 3 (‘Empathy queens’), para. 5.

7
(Eisenberg & Lennon, 1983).

8
Quoted in (Schaffer, 2008), entry 3 (‘Empathy queens’), para. 8.

9
(Davis & Kraus, 1997), p. 162.

10
(Ames & Kammrath, 2004), p. 205; (Realo et al., 2003), p. 434.

11
(Voracek & Dressler, 2006).

12
Both the EQ and the Reading the Mind in the Eyes test, also from Simon Baron-Cohen’s lab, ask participants to state their sex before beginning the
questionnaire. As will become clear later in the chapter, it’s possible that the correlation between the two arises because the salience of gender-related norms increases both self-reported empathy and empathic performance, to a greater or lesser degree in different participants.

13
(Ickes, 2003), p. 172.

14
(Levy, 2004), p. 322.

15
(Voracek & Dressler, 2006). If you used information about whether someone scored below or above average on the test to try to guess his or her sex you would be correct barely more often than chance.

16
These and further details of the PONS and its interpretation, as well as the IPT, are summarised in (Graham & Ickes, 1997). To give you an idea of the size of the gender difference on the PONS, which Graham and Ickes describe as ‘respectable’ (p. 123), the average woman on this test (scoring at the 50th percentile) is equivalent to a slightly superior man (scoring at the 66th percentile for the male population). In their discussion of the greater female advantage for ‘leaky’ channels of communication, they are referring to the work (and term) of R. Rosenthal and B. DePaulo, ‘Sex differences in eavesdropping on nonverbal cues’,
Journal of Personality and Social Psychology
37 (1979), pp. 273–285.

17
(Brizendine, 2007), p. 160.

18
This hypothesis again refers to the work of Rosenthal & DePaulo, cited in (Graham & Ickes, 1997).

19
(Graham & Ickes, 1997), p. 126.

20
(Ickes, 2003), quotations from pp. 125 and 126, respectively.

21
(Ickes, Gesn, & Graham, 2000).

22
(Ickes, 2003), p. 135.

23
(Klein & Hodges, 2001). Men also scored equivalently to women when the sympathy rating was requested
after
the empathic accuracy test.

24
(Thomas & Maio, 2008), p. 1173. This effect was only found for an easy-to-read target, not a difficult-to-read target.

25
(Koenig & Eagly, 2005), p. 492.

26
(Marx & Stapel, 2006c), p. 773.

27
(Seger, Smith, & Mackie, 2009), p. 461.

28
(Ryan, David, & Reynolds, 2004). Gilligan’s work and critiques summarised here also.

29
This claim also found support in (Ryan, David, & Reynolds, 2004), study 1.

30
(Ryan, David, & Reynolds, 2004), pp. 253 and 254, respectively, references removed.

1
For meta-analysis, see (Voyer, Voyer, & Bryden, 1995).

2
(Moore & Johnson, 2008; Quinn & Liben, 2008). It’s worth noting that the early appearance of this difference does not necessarily mean that experiential factors could not be responsible. For example, male babies could be given more gross stimulation that stimulates visuospatial skills. Interestingly, one study found that boys and girls from a low socioeconomic background underperformed equally on a visuospatial task, whereas more-privileged boys outperformed their female counterparts. This points towards the importance of experiential factors in male advantage (Levine et al., 2005). Moreover, an early advantage for males doesn’t mean that this must inevitably persist. In other cognitive domains, gender differences are transient.

3
Needless to say, this is a complex issue. As Nora Newcombe recently summarised it, not only do men, on average, outperform women on mental rotation tasks, particularly at the highest levels, spatial visualisation skills are relevant to success in fields such as physics, mathematics, computer science and engineering. However, as she also notes, there are difficulties with the argument that these genuine sex differences are biologically caused and immutable. With regard to the first point – biological causation – she notes that hypotheses attempting to account for biological mechanisms have not been successful. (Two of these, hormonal accounts and sex differences in lateralisation, are discussed in the second part of the book. The other ideas – an X-linked recessive gene for spatial ability and males’ later puberty – have not been supported by the evidence.) Newcombe also notes that, despite superficial plausibility, evolutionary explanations entail numerous untested assumptions. One further important point raised by Newcombe is whether extra increments in mental rotation ability are important, beyond some high threshold. (As Amanda Schaffer dryly put it in
Slate
, ‘when it comes to the diverse precincts of high-level science, spatial reasoning only gets you so far. Rock-star academics don’t necessarily spend their days turning geometric figures around in their minds.’) Newcombe points out that ‘[t]hinking creatively, explaining one’s data, or inspiring a research team may be pretty important as well!’ (Newcombe, 2007), p. 75. A recent, very comprehensive review of ‘sociocultural and biological considerations’ with respect to women’s underrepresentation in science concluded that the ‘process needed to establish male advantage in STEM fields as a function of superior spatial ability (possibly because of its role in advanced mathematics) is littered with loopholes. Nothing close to a tightly reasoned and supported argument currently exists.’ (Ceci, Williams, & Barnett, 2009), p. 250.

4
Reviewed, for example, by (Newcombe, 2007). Recent studies have also found that playing computer games improves mental rotation ability, and in women more so than in men (Cherney, 2008; Feng, Spence, & Pratt, 2007).

5
(Sharps, Price, & Williams, 1994). Task instructions quoted from pp. 424 and 425. Men in the masculine condition outperformed men and women in all other groups. See also (Moè & Pazzaglia, 2006).

6
(McGlone & Aronson, 2006).

7
(Hausmann et al., 2009).

8
(Moè, 2009).

9
M. B. Ritter,
More than gold in California 1849–1933
(Berkeley, CA: The Professional Press, 1933), p. 161. Quoted in (Morantz-Sanchez, 1985), p. 118.

10
C. M. Steele, S. J. Spencer, & J. Aronson, ‘Contending with group image: The psychology of stereotype and social identity threat’. In M. P. Zanna (ed.),
Advances in experimental social psychology
, vol. 34 (San Diego: Elsevier, 2002), p. 385. Quoted in (Shapiro & Neuberg, 2007), p. 109.

11
Readers interested in reading more about stereotype threat are strongly recommended to visit the Web site
http://reducingstereotypethreat.org
, authored by social psychologists Steven Stroessner and Catherine Good, which provides detailed and comprehensive coverage of the academic literature.

12
(Good, Aronson, & Harder, 2008).

13
For example (Marx & Stapel, 2006b; Marx, Stapel, & Muller, 2005; Thoman et al., 2008).

14
(Good, Aronson, & Harder, 2008), p. 25.

15
(Walton & Spencer, 2009), p. 1133. Although they note that their samples may not be representative of the general population, their effect sizes suggest that the SAT Maths may underestimate women’s abilities by about 20 points (compared with a gender gap of 34 points). For African and Hispanic Americans, SAT Reading tests may underestimate ability by about 40 points.

16
For example (Adams et al., 2006; Danaher & Crandall, 2008; Davies et al., 2002; Inzlicht & Ben-Zeev, 2000; Logel et al., 2009).

17
See (Nguyen & Ryan, 2008).

18
(Marx, Stapel, & Muller, 2005).

19
For example (Cadinu et al., 2003; Stangor, Carr, & Kiang, 1998) and (Marx & Stapel, 2006a), p. 244. As David Marx has argued, and his work has been demonstrating, priming a self-relevant stereotype has effects different from, and greater than, standard stereotype priming effects.

20
(Cadinu et al., 2005), p. 574.

21
(Logel et al., 2008). See also (Davies et al., 2002) who found that gender stereotypes were activated in women who saw gender-stereotyped
advertisements, compared with controls, and that this activation predicted maths underperformance.

22
(Beilock, Rydell, & McConnell, 2007; Schmader & Johns, 2003). For review see (Schmader, Johns, & Forbes, 2008).

23
(Johns, Inzlicht, & Schmader, 2008).

24
For example (Aronson et al., 1999; Croizet et al., 2004).

25
Presenting the test as gender-neutral (i.e., males and females score equally) enhances women’s performance (for example [Spencer, Steele, & Quinn, 1999]), and does not have the same detrimental effect on working memory (for example [Johns, Inzlicht, & Schmader, 2008]).

26
(Seibt & Förster, 2004).

27
(Gladwell, 2008), pp. 87, 87, and 88, respectively.

28
See (Nguyen & Ryan, 2008) who concluded from their meta-analysis that low maths-identified women are the least affected by stereotype threat. Interestingly, they found that moderately identified women are the most affected (more so than high-identified females), although they note that there is some inconsistency in how ‘identification’ is defined and operationalised.

29
For instance (Beilock, Rydell, & McConnell, 2007) found that stereotype threat most affects maths problems that rely more heavily on working-memory resources.

30
These numbers, from the National Science Foundation, are cited in (Ceci, Williams, & Barnett, 2009).

31
(Inzlicht & Ben-Zeev, 2000).

32
(Schmader, Johns, & Barquissau, 2004).

33
(Kiefer & Sekaquaptewa, 2007).

34
See (Blanton, Crocker, & Miller, 2000; Marx, Stapel, & Muller, 2005). For effect of ‘closeness’ of the model, see (Marx et al., unpublished manuscript), who found that women exposed to a highly maths-competent, socially ‘close’, female role model performed better on a maths test than women exposed to a socially ‘distant’, but equally competent, female role model. (Lockwood, 2006) found that women in particular benefit by having an inspiring female role model. In general, research into social comparison processes finds that, among other factors, our self-evaluations and behaviour are more likely to assimilate to another person to the extent that we feel psychologically similar to them. Otherwise the standard set by the other person becomes a contrast against which our own self-evaluation and behaviour reacts. See, for example (Mussweiler, Rüter, & Epstude, 2004).

35
(Marx & Roman, 2002; McIntyre et al., 2005; McIntyre, Paulson, & Lord, 2003).

36
(Josephs et al., 2003; Newman, Sellers, & Josephs, 2005).

37
See (Rogers, 1999), pp. 75–85. It’s also worth noting that although some have
argued that the relationship between testosterone levels and competition is different in women and men, there are currently too few studies available with women to draw a fair comparison. See (van Anders & Watson, 2006), pp. 215–220.

38
(Sherwin, 1988).

39
(Rogers, 1999), p. 83.

40
(Josephs et al., 2003), p. 162.

41
(Huguet & Régner, 2007; Neuville & Croizet, 2007). Also (Ambady et al., 2001) found stereotype threat effects in lower-elementary and middle school girls, although unexpectedly upper-elementary girls did better when gender identity was salient.

42
(Nosek et al., 2009), p. 10597. These relationships held even controlling for a general indicator of social gender inequality.

1
(Hines, 2004), p. vii.

2
(Haslanger, 2008), p. 211.

3
Quoted in (McCrum, 2008) p. 22.

4
(Mullarkey, 2004), pp. 369 and 370, respectively.

5
(Mullarkey, 2004), pp. 373 and 374.

6
(Pinker, 2008), p. 5.

7
(Steele, 1997), p. 618.

8
(Murphy, Steele, & Gross, 2007).

9
(Davies et al., 2002).

10
(Davies, Spencer, & Steele, 2005).

11
(Gupta & Bhawe, 2007), p. 74.

12
A point made by (Cheryan et al., 2009).

13
(Cheryan et al., 2009).

14
I. J. Seligsohn,
Your Career in Computer Programming
(New York: Simon & Schuster, 1967), cited in (Gürer, 2002a), p. 176.

15
(Gürer, 2002b), p. 120.

16
Sapna Cheryan, personal communication, November 25, 2009.

17
(Cheryan et al., 2009), p. 1058.

18
(Spelke & Grace, 2006), p. 726.

19
The criteria were changed to downplay prior programming ability – which was shown not to be a predictor of success in the CS major, and instead focus on ‘indicators of future visionaries and leaders in computer science.’ (Blum & Frieze, 2005), p. 117. The study referred to was conducted by Jane Margolis
and Allen Fisher, reported in
Unlocking the Clubhouse
(Cambridge, MA: MIT Press, 2002).

20
(Blum & Frieze, 2005), quotations from pp. 113 and 114.

21
(Good, Rattan, & Dweck, unpublished).

22
(Haslanger, 2008), p. 212.

23
(Correll, 2001), p. 1724.

24
(Correll, 2004), p. 102.

25
(Pronin, Steele, & Ross, 2004).

26
The article was C. P. Benbow and J. C. Stanley, ‘Sex differences in mathematical ability: fact or artifact?’
Science
210 (1980), pp. 262–1264.

27
Quoted in (Pronin, Steele, & Ross, 2004), p. 159.

28
(Hewlett, Servon et al., 2008), p. 11.

29
(Hewlett, Luce, & Servon, 2008), p. 114.

30
(Hewlett, Servon et al., 2008), quotations from pp. 11 and 12.

31
For instance, from their comprehensive review of possible biological and social factors contributing to female underrepresentation in science, Stephen Ceci and colleagues conclude that the evidence for the role of biological factors is ‘contradictory and inconclusive.’ They suggest that the evidence points most strongly to the role of women’s preferences – which they note could either be seen as free choices or constrained ‘choices’ – with a secondary factor being poorer performance on gatekeeper tests, which they regard as being more likely due to sociocultural than biological factors (Ceci, Williams, & Barnett, 2009), p. 218.