Issues in Psychology

Throughout your studies certain issues have been considered repeatedly: gender, culture, and ethics. In this chapter you have the opportunity to reflect on these issues in relation to your studies so far.

Ask yourself

Is it justifiable if a theory is based on male behaviour only and then applied to all human behaviour?
Which is preferable: to regard men and women as being psychologically different, or as the same?
How can we make comparisons between cultures without being biased by our own cultural assumptions?
What do psychologists think about the concept of “race”?
How can we balance the needs of the individual against the needs of society?
Do the ends justify almost any means?
Is current legislation sufficient to protect non-human animals?
What are the differences between humans and non-human animals?

GENDER BIAS

The term “bias” is used to suggest that a person’s or society’s views are distorted in some systematic way. In psychology, there is evidence that gender is presented in a biased way and this bias leads to a misrepresentation of women. Consider the following example. The performance of participants in psychological research tends to be influenced by the expectations of the investigator. Many people still have lower expectations for women than for men. This would lead us to collect data that show poorer task performance in women (for example, on a memory task). Research data are used to formulate theories and these theories may well be gender biased because of the baseline data. In this Section we will explore different areas and aspects of gender bias, and how this affects psychological knowledge.

Gender Stereotypes

There are many popular (and misleading) stereotypes about the differences between the sexes. For example, it has often been claimed that women are more emotional than men. This was expressed poetically by Alfred, Lord Tennyson:
Man for the sword and for the needle she: Man with the head and woman with the heart.
It is fully accepted that that is a gross oversimplification of reality. However, it is worth mentioning that females typically score significantly higher than males on measures of negative affectivity (a personality dimension relating to the experience of negative emotions such as anxiety and depression) (e.g., Denollet, 2005).
Stereotypes about gender have been fairly common in psychology as well as in society at large. One of the worst offenders was Sigmund Freud. He argued that anatomy is destiny, meaning that there are great psychological differences between men and women because of their anatomical differences. For example, Freud claimed that young girls suffer from “penis envy” when they find out that boys have a penis but they don’t.
The greatest difficulty lies in distinguishing “real” from culturally created gender differences. There are some real differences, or at least that was the conclusion reached by Maccoby and Jacklin (1974) in a review of research on sex differences (see Eysenck’s A2 Level Psychology Chapter 7, Gender). They concluded that there were only four differences between boys and girls for which there was strong evidence. This is a much smaller number of gender differences than would have been predicted by most psychologists. The four differences identified by Maccoby and Jacklin were as follows:

Girls have greater verbal ability than boys.
Boys have greater visual and spatial abilities than girls (e.g., arranging blocks in specified patterns).
Boys have greater arithmetical ability than girls, but this difference only appears at adolescence.
Girls are less aggressive than boys: this is found in nearly all cultures, and is usually present from about 2 years of age.

Most of these differences are fairly small, and there is much overlap in behaviour between boys and girls. Sex differences in abilities (verbal, visual, spatial, and mathematical) are even smaller now than they were in the early 1970s (Hyde & Linn, 1988). However, there are clear (and increasing) gender differences in academic performance in the UK. For example, findings contained in a 2005 report indicated that girls outperformed boys at A level in 74% of state schools, a difference that probably reflects gender differences in motivation.
As Shaffer (1993) pointed out, there are some differences that were not identified by Maccoby and Jacklin (1974). First, girls show more emotional sensitivity (e.g., they respond more attentively to babies). Second, girls are less vulnerable developmentally than boys and are less likely to suffer from learning disabilities, various language disorders, or hyperactivity. Third, boys tend to be more physically active than girls.
In a large-scale survey of gender stereotypes in 30 different national cultures, Williams and Best (1982) found that there were many similarities across the various cultures. Men were seen as more dominant, aggressive, and autonomous; a more instrumental role. Women were more nurturant, deferent, and interested in affiliation; being encouraged to develop an expressive role. This suggests some kind of universal, biological basis for gender stereotypes. However, we mustn’t assume that any of these differences are always found. Whyte (1978) considered 93 non-industrialised cultures, and found that men dominated their wives in 67% of those cultures. There was equality of the sexes in 30% of the cultures and women dominated their husbands in the remaining 3%. Unsurprisingly, men were particularly dominant in non-industrialised cultures in which they were better equipped than women to perform tasks vital within that culture (e.g., cultures engaging in frequent warfare; Goldstein, 2001).

Alpha Bias and Beta Bias

If there are real gender differences, how does that affect psychological research? Hare-Mustin and Marecek (1988) considered the issue of gender bias in psychology in detail. Their starting point was that there are two basic forms of gender bias: alpha bias and beta bias. According to Hare-Mustin and Marecek (p.457), “Alpha bias is the tendency to exaggerate differences; beta bias is the tendency to minimise or ignore differences.” They used the term “bias” to refer to an inclination to focus on certain aspects of experience rather than on others.
Febbraro (2003) pointed out that we can see these biases in the work/family literature. Alpha bias is involved when it is argued that women experience much greater work/family stress than men, and so male-dominated structures need to be transformed as a result. Beta bias is involved when it is argued that multiple roles (i.e., parent, spouse, worker) increase the well-being of women in the same way as men.

Alpha bias

Within Western cultures, alpha bias has been more common than beta bias. For example, Freud claimed that children’s superego or conscience develops when they identify with the same-sex parent. Girls don’t identify with their mother as strongly as boys identify with their father. As a result, Freud argued that girls develop weaker superegos than boys. However, Freud did admit that “the majority of men are far behind the masculine ideal [in terms of strength of superego]”. The evidence doesn’t support Freud. Hoffman (1975) discussed studies in which the tendency of children to do what they had been told not to do was assessed. The behaviour of boys and girls did not differ in most of the studies. When there was a sex difference, it was the girls (rather than the boys) who were better at resisting temptation.
The approach adopted by evolutionary psychologists has often been criticised for its alpha bias. According to evolutionary psychologists, evolutionary processes in the development of the human species explain why men tend to be dominant over women, why women typically have much more parental investment than men in their offspring, and why men are more likely to commit adultery. However, there have been major cultural changes over the past 100 years or so, as a result of which it is increasingly argued that the evolutionary perspective shouldn’t be used to justify gender differences.
There is some evidence for alpha bias in the diagnosis of mental disorders. Ford and Widiger (1989) argued that histrionic personality disorder (characterised by excessive emotionality) is often regarded as a distortion of stereotypical feminine traits, whereas anti-social personality disorder (characterised by hostility and aggression) is a distortion of stereotypical masculine traits. Therapists were presented with written case studies of histrionic and antisocial personality disorder in which the patient was identified as male or female. Histrionic personality disorder was diagnosed much more frequently when the patient was female, and anti-social personality disorder was diagnosed more often when the patient was male.
You won’t be surprised to discover that the great majority of examples of alpha bias are biased in the direction of favouring males over females. However, there are a few cases in which females are regarded as superior to males. Gilligan (1977, 1982) argued that the morality of care is very important and that women adopt this form of moral reasoning to a greater extent than men. It should be noted that the evidence provides little support for Gilligan’s views (e.g., Jaffee & Hyde’s, 2000, meta-analysis). However, as Durkin (1995, p.493) pointed out, Gilligan’s “critical perspective did serve the purpose of opening up the study of moral development in important ways by broadening conceptions of what morality is and how it should be measured”.
There is a final point that needs to be made about alpha bias. Most journals in psychology are much more willing to publish significant findings than non-significant ones. As a result, it is likely that the published literature exaggerates the extent of gender differences in behaviour—researchers finding no gender differences find it difficult to publish their data.

Beta bias

Hare-Mustin and Marecek (1988) argued that beta bias, or the tendency to minimise or ignore sex differences, is less common than alpha bias. They suggested that Bem’s (1974) theory of psychological androgyny (see A2 Level Psychology Chapter 7, Gender) is an example of beta bias. According to that theory, it is psychologically more healthy to be androgynous (having a mixture of positive masculine and feminine characteristics) than to have only masculine or only feminine characteristics. Individuals who can respond to any situation with either masculine (instrumental) characteristics or feminine (expressive) characteristics are more flexible than an individual who behaves in a more sex-stereotyped way.

Ask yourself: Why is alpha bias more common than beta bias?

There is evidence of beta bias in experimental research, i.e., a tendency to reduce or minimise gender differences. Male and female participants are used in most studies, but there is typically no attempt to analyse the data to see whether there are significant sex differences. It may be possible that sex differences are found in psychological research because researchers ignore the differential treatment of participants. Male experimenters may treat their female participants differently from their male ones. Rosenthal (1966) reported that they were more pleasant, friendly, honest, and encouraging with female than with male participants. Such findings led Rosenthal to conclude: “Male and female subjects may, psychologically, simply not be in the same experiment at all.”

Ask yourself: How might an experimenter control for this differential treatment of men and women?

This means that, because researchers act as if there are no sex differences, they end up providing evidence that there are such differences. In other words beta bias tends to produce sex differences.

Beta bias in psychological theories

The same reasoning can be applied to some psychological theories that show evidence of beta bias. Kohlberg (1963) put forward a theory of moral development based mainly on studies of moral dilemmas with males as the main actors and with males as participants. He claimed that men tended to be at a higher level of moral development than women (see A2 Level Psychology Chapter 9, Cognition and Development). Kohlberg assumed that there were minimal differences between men and women in terms of moral thinking (a beta bias) and therefore it would not matter if he used only male participants because this would still represent all people. The outcome is a demonstration of gender differences.

Ask yourself: How would you measure moral development in a way that is not gender specific?

An androcentric view of human nature

The result of beta bias in psychological research is that we end up with a view of human nature that purports to apply to men and women alike, but in fact has a male or androcentric bias. This is true of Kohlberg’s theory of moral development. It is also true of other areas of research. Asch’s (1955) conformity studies involved all male participants. In fact, the same is true of many of the other conformity studies (e.g., Perrin & Spencer, 1980). Psychological theories of conformity are thus based largely on male behaviour. Eagly (1978) found that women may be even more conformist, or at least they are more oriented towards interpersonal goals and therefore appear to be more conformist in experimental situations. What is clear is that a failure to attend to gender issues has oversimplified, exaggerated, and/or fudged gender differences!

Ask yourself: Moscovici et al.’s (1969) classic study on minority influence used all female participants. If he had used male participants, do you think minority influence would have appeared to be less significant?
Ask yourself: Sherif et al.’s (1961) study of boys at a summer camp involved all boys, and Erikson’s (1968) research into lifespan development also involved all male interviews. What other studies used all male or all female participants? How may this have affected the theories derived from the data?

Facts and Values

The discussion so far has assumed that there are facts about gender. A different view is favoured by social constructionists such as Gergen (1985). They argue that there are no facts; there are values and these determine what are regarded as facts. We construct our reality through shared conversations about the social and physical world. In other words, “scientific knowledge, like all other knowledge, cannot be disinterested or politically neutral” (Hare-Mustin & Marecek, 1988, p.456). Thus we may be “fooled” into believing that our knowledge of gender facts is “real”, whereas they are social constructions.
Such facts may have serious repercussions. Burns (1993) provided an example of how society’s values can influence the approach taken to studying women. She pointed out (p.103) that a major focus of research on women with learning disabilities is on “sexuality and the issues and concerns surrounding women with learning disabilities becoming pregnant, having babies, being sterilised, using contraception, managing periods and being sexually abused”. Thus, such women are seen in a negative way in terms of the possible problems they may cause. As Burns (p.103) concluded, “the consequence of this position is to deny women with learning disabilities a positive identity and role as a woman”.

Methodological gender bias

We have already seen that there are various ways in which research on gender differences can be biased and thus invalid. Another way in which findings can be rendered invalid is through methodological gender bias, in which the design of the research biases the chances of the researcher obtaining some particular finding. Methodological gender bias is especially likely to be found when the direction of gender differences depends on the precise measures of behaviour that are taken. For example, many researchers have addressed the issue of whether males or females are more aggressive. In one of the most thorough studies, Bjorkqvist et al. (1992) found that boys displayed much more physical aggression than girls, but girls showed significantly more indirect aggression (e.g., gossiping, writing unkind notes) than boys. Armed with that knowledge, you could design a study to show apparently that boys are more aggressive than girls or vice versa!
Eagly et al. (1995) carried out a meta-analysis to compare the effectiveness of male and female leaders. They found that there was overall very little difference between male and female leaders, with female leaders being more effective than male leaders in 57% of significant comparisons and male leaders being more effective than female leaders in 43%. The key finding was that female leaders were consistently more effective than males in feminine leader roles (i.e., roles rated as being congenial to women) and male leaders were more effective than females in masculine leader roles (i.e., roles rated as being congenial to men). Thus, the results depended very much on the situation—you could apparently show that men make better leaders if you focused on leadership in the army, but could show that women make better leaders if you focused on leadership among nurses!
In sum, it is important for researchers studying gender differences to make sure that they compare the two sexes in a range of different situations and using various measures of behaviour. If that is not done, then there is a very real risk of methodological gender bias.
Activity: Facts, values, and gender