7
Social Comparison: Theory, Research, and Applications
Jan Crusius, Katja Corcoran, and Thomas Mussweiler
Social comparisons—comparisons between the self and others—are a fundamental mechanism influencing people’s judgments, experiences, and behavior. Psychological research supports the notion that people constantly engage in social comparisons. Arguably, whenever they receive information about how others are, what others can and cannot do, or what others have achieved and have failed to achieve, they are inclined to relate this information to themselves (Dunning & Hayes, 1996). Likewise, whenever they want to know how they themselves are or what they themselves can and cannot do, they are likely to do so by comparing their own characteristics, fortunes, and weaknesses to those of others. One indicator for this robustness of social comparison is that people may sometimes even engage in comparisons with others who do not yield relevant information concerning the self (Gilbert et al., 1995). Another sign of the importance of social comparisons is their power in eliciting universal human emotions. We may feel pride when we succeed in outperforming competitors, marvel in admiration about the excellence of other but may also feel the pain of envying them (Crusius & Lange, 2017; Smith, 2000; Steckler & Tracy, 2014).
Because comparisons with others are such an essential human proclivity, it may not be surprising that social comparison is a highly studied topic within social psychology. Three broad questions have guided this research: Why do people engage in social comparisons? To whom do they compare themselves? How do social comparisons influence the self?
Festinger’s Social Comparison Theory
Researchers have investigated these questions with a wide variety of theoretical approaches over the last decades. Nevertheless, in several ways, these questions origin from Leon Festinger’s seminal publications in the 1950s. In the Lewinian tradition, Festinger was interested in how forces of the environment affect individuals’ opinions leading to conformity. Eventually leading to his work on social comparisons was his research on communication processes within groups (Festinger, 1950). He argued that communication promotes agreement in groups, which is required because of two reasons. First, group members should hold similar opinions to facilitate group cooperation and achievement. Second, to the extent that group consensus validates the accuracy of one’s opinions and preferences, it serves the need for a shared social reality. Thus, in his theorizing about social communication, Festinger stressed the importance of others in the formation of one‘s opinions.
In his theory of social comparison (Festinger, 1954), he added the realm of abilities. He emphasized that individuals attend to others to fulfill their own need to gain knowledge about themselves, not only to gain insight into how their beliefs relate to those of others but also to ascertain what their abilities enable themselves to achieve.
Festinger’s theory of social comparison builds upon nine hypotheses (Table 7.1). That his interest in social comparison emerged from his work on opinions is apparent in several of them (e.g., Hypotheses I to V). Yet, very little research on comparisons of opinions followed. Most of the subsequent research has concentrated on comparisons of abilities (e.g., Wheeler, 1966) and domains that are predictors of abilities, such as personality traits (e.g., Hakmiller, 1966), or consequences of abilities such as emotions in response to threat (e.g., Wrightsman Jr., 1960). Furthermore, this focus on ability comparisons prompted attention toward precursors, processes, and consequences of comparisons at the individual level. In contrast, how comparisons are determined by and shape outcomes at the collective level was largely neglected (for a review, see Križan & Gibbons, 2014), even though this question was Festinger’s starting point.
Table 7.1 Hypotheses of the theory of social comparison processes (Festinger, 1954).
No.Hypothesis
IThere exists, in the human organism, a drive to evaluate his opinions and his abilities.
IITo the extent that objective, non–social means are not available, people evaluate their opinions and abilities by comparison respectively with the opinions and abilities of others.
IIIThe tendency to compare oneself with some other specific person decreases as the difference between his opinion or ability and one’s own increases.
IVThere is a unidirectional drive upward in the case of abilities which is largely absent in opinions.
VThere are non-social restraints that make it difficult or even impossible to change one’s ability. These non-social restraints are largely absent for opinions.
VIThe cessation of comparison with others is accompanied by hostility or derogation to the extent that continued comparison with those persons implies unpleasant consequences.
VIIAny factors which increase the importance of some particular group as a comparison group for some particular opinion or ability will increase the pressure toward uniformity concerning that ability or opinion within that group.
VIIIIf persons who are very divergent from one’s own opinion or ability are perceived as different from oneself on attributes consistent with the divergence, the tendency to narrow the range of comparability becomes stronger.
IXWhen there is a range of opinion or ability in a group, the relative strength of the three manifestations of pressures toward uniformity will be different for those who are close to the mode of the group than those who are distant from the mode. Specifically, those close to the mode of the group will have stronger tendencies to change the positions of others, relatively weaker tendencies to narrow the range of comparison and much weaker tendencies to change their position compared to those who are distant from the mode of the group.
One additional reason for this pronounced shift may have been that Festinger himself did not linger for long with his theory of social comparison processes but quickly moved on to formulate the theory of cognitive dissonance (Festinger, 1957). Notwithstanding, his theory inspired and shaped the research on social comparison for decades to come. Coming back to the three central questions, his hypotheses offer several initial propositions for answers.
In particular, Hypotheses I and II speak to why people engage in social comparisons. Festinger points out that having “incorrect opinions and/or inaccurate appraisals of one’s abilities can be punishing or even fatal in many situations” (Festinger, 1954, p. 117). Thus, he proposed that it is essential to need to know the self. However, it should often be impossible to determine opinions or abilities by reference to the physical world. In combination, these factors should motivate people to compare themselves to other people.
Hypotheses III, IV, and VIII summarize Festinger’s understanding of the standards with whom people will compare. In essence, he postulated that people will seek out similar others for comparisons, or, in the case of abilities, others who are slightly better. He argued that comparisons with people whose opinion or ability are too discrepant do not provide much useful information for assessing the accuracy of one own opinion or ability—mostly because the result of such a comparison is known beforehand.
Finally, Hypotheses V, VI, VII, and IX point out consequences of social comparisons to the self. Comparisons might cause a change in one’s opinion or ability and promoting uniformity. In other words, Festinger presumed a dominance of assimilation as an outcome of social comparison. Furthermore, he theorized that the amount of change would greatly depend on the importance, relevance, and attraction to the comparison group and that the inability to reach uniformity would be perceived as unpleasant.
Thus, Festinger touched upon reasons for social comparison, the relevance of comparison standards, and the consequences of social comparisons. Subsequent research investigated some of his propositions, but also branched out to examine these guiding themes with diverging hypotheses.
Why Do People Engage in Social Comparisons?
The classic answer for why people compare themselves to others rests on motivational considerations. As outlined above, in the original theory of social comparisons, Festinger (1954) stressed the desire of people to know themselves. He postulated that people have a basic need to maintain stable and accurate views of the self. Therefore, they should be inclined to seek informative feedback about their characteristics and abilities. Festinger proposed that people mainly rely on objective standards for such evaluations. However, objective standards might not always be available, or comparisons with them can be hard to achieve. In this case, people should fall back on social comparisons with others.
Later research questioned this preference for objective standards. Instead, people appear to compare with others even if objective standards are present (Klein, 1997). Relatedly, neuroscientific evidence suggests that social comparisons can have a greater impact than equivalent information about absolute outcomes (Bault et al., 2011; Kedia et al., 2014). Furthermore, even when people have highly diagnostic, aggregate data about many others, the impact of this information can be relatively small. Instead, comparisons with singular individuals that are present in the closer social environment (such as friends or peers) or in the immediate situation can have a greater influence on self-evaluation (Gerber et al., 2018; Marsh & Parker, 1984; Zell & Alicke, 2010) and well-being (Anderson et al., 2012). Zell and Alicke (2010) argue that the reason for this local dominance effect is that our minds have evolved in small groups. Knowing how we compare to our close social competitors may have been of utmost importance in our evolutionary past. On a more general level, these considerations align with evolutionary theorizing that humans are characterized by a strong proclivity to precisely monitor social rank as a fundamental part of their pursuit of social status (Anderson et al., 2015). In any case, one important motive for social comparisons seems to be the need for accurate self-evaluations (for an overview, see Taylor et al., 1996), even if trying to gain such insights about the self can be fraught with biases.
However, it is also clear that, at least sometimes, people do not seek accurate feedback about themselves. Instead, they may try to create and maintain a positive self-image. One important means to do so is to compare with others who are worse-off. Even failures might suddenly appear to be successes in comparisons with other people who performed even worse. Thus, such downward comparisons may be an effective way to protect the self (Wills, 1981).
A third need eventually fulfilled by social comparison may be to self-improve (Taylor & Lobel, 1989). To gain information and hints on how to advance and improve skills, people should seek comparisons particularly with upward standards—others, who are better than they are.
Thus, social comparisons are typically portrayed as strategic processes, which are executed to satisfy certain motives or goals (Taylor et al., 1996). Specifically, social comparison is mostly understood as a purposeful psychological tool fulfilling fundamental needs like self-evaluation, self-enhancement, and self-improvement (Kruglanski & Mayseless, 1990; Suls et al., 2002; J. V. Wood & Taylor, 1991). However, not all social comparisons appear to be such a deliberate and strategic process. On the contrary, as documented by a long history of experimental research on social comparisons (for a meta-analysis, see Gerber et al., 2018), comparisons can be forced upon people by the situation that they happen to be in. Furthermore, social comparisons may be conducted spontaneously and without intention (Gilbert et al., 1995; Mussweiler et al., 2004a). In this case, they might not be strategic means to fulfill a certain goal. Such spontaneous comparisons suggest additional reasons why people compare with others.
One of these reasons arises from logical and conversational considerations. Information about characteristics, abilities, and achievements often concern attributes and dimensions that are defined in a relative manner. To say that somebody is athletic, intelligent, or attractive implies that this person is more athletic, more intelligent, and more attractive than others (Huttenlocher & Higgins, 1971). Therefore, it is necessary to compare the target with relevant standards to interpret incoming information correctly and to communicate outgoing information successfully. If one, for example, hears the simple statement “Tom is tall,” one would estimate his body height differently if Tom were a toddler versus a basketball player. To understand the statement correctly, one needs to know the reference point and its relative meaning. Research in social psychology has repeatedly demonstrated that people usually consider their communication partners’ norms and standards while processing or exchanging social information (Biernat & Manis, 1994; Grice, 1975; Schwarz, 1994).
A further answer to the question of why people engage in social comparisons may be that social comparisons are cognitively efficient (Corcoran & Mussweiler, 2010; Keil et al., 2006; Mussweiler & Epstude, 2009). Specifically, social comparisons may allow to process information in a more efficient manner than absolute modes of information processing. As people are cognitive misers with limited cognitive processing power (Taylor, 1981), they may be inclined to resort to social comparisons as a useful tool to save such scarce cognitive resources. Specifically, comparisons limit the range of information that is relevant to evaluate or judge a given object. For example, if you try to judge your athletic abilities, considering all the information that has some implication for this judgment would be an arduous endeavor. Instead, you could relate your athletic abilities to context-relevant comparison standards. If these standards prescribe running and cycling abilities as relevant dimension but not ski or swim, you would only consider the former information about yourself and not the latter. Research by Mussweiler and Epstude (2009) supports this notion. Participants induced to make judgments in a comparative rather than absolute manner relied on less information, were faster in making critical judgments, and had more resources available for the processing of a secondary task.
Taken together, there appear to exist several reasons why people compare themselves to others. On the one hand, people might want to satisfy important needs like self-evaluation, self-enhancement, and self-improvement (Festinger, 1954; Taylor & Lobel, 1989; Taylor et al., 1996; Wills, 1981). On the other hand, the reference to comparison standards might be necessary to communicate successfully with others (Huttenlocher & Higgins, 1971; Schwarz, 1994). Finally, social comparisons might be an efficient cognitive tool to gain self-knowledge without binding to many cognitive resources (Corcoran & Mussweiler, 2010; Mussweiler & Epstude, 2009).
Chapter 8
8
Relative Deprivation and Cognate Theories: Making Sense of Irrational Behaviors
Beverly G. Conrique and Faye J. Crosby
“A house may be large or small; as long as the neighboring houses are likewise small, it satisfies all social requirements for a residence. But let there arise next to the little house a palace, and the little house shrinks to a hut.”
– Karl Marx, 1847/1935 , p. 33
Introduction
On November 8, 2016, the United States astonished the world and elected to its highest office a man who had never held elected position, who had openly disrespected women, people of color, people with disabilities, and immigrants (Neate, 2015; Trump, 2015), and who had put the interests of a small cadre of wealthy supporters ahead of national interests (Gold & Narayanswamy, 2016). Minute-to-minute reporting of the election that night carried with it undertones of astonishment and disbelief (Arkin & Siemaszko, 2016; Jensen, 2016; Sullivan, 2016). Although many attributed Donald Trump’s win to anti-female bias (Clinton, 2017; Crosby, 2017), subsequent analysis has shown that other factors must also have come into play.
Traditionally, in the United States and elsewhere, people align their votes with their financial and other interests. While the 2016 election did not totally deviate from the traditional pattern (Scott, 2017, 2018), some odd phenomena occurred. During the course of the campaign, a union of immigration and enforcement officers – working-class, blue-collar individuals – unexpectedly announced their official support for Trump (Bacon, Jr., 2017), and exit polls suggested nearly 50% of union households in Ohio voted for Trump (Greenhouse, 2016). In the past, unions backed only Democrats.
American blue-collar support for a blustery billionaire from New York City in 2016 calls to mind how blue-collar workers associated themselves during the mid-20th century with the Tory party in England. That connection, in turn, might make social scientists think of the concept of relative deprivation (RD), a concept that scholars like Runciman (1966) have used to explain how people of modest means might align with conservatives whose policies aim to benefit the well-off, not the poor.
Relative deprivation – as a concept and as a theory – is at the heart of this chapter. The term “relative deprivation” dates back to research conducted by Samuel Stouffer and colleagues during World War II (Stouffer et al., 1949). In their study of the American soldier, Stouffer and colleagues discovered that military police, despite their slower promotions, felt happier with their promotion rates compared to other military personnel, who were promoted more frequently (Stouffer et al., 1949). They also found that Black soldiers from the South were measurably happier with life in the Army than Black soldiers from the North, presumably because of differences in their lives outside the Army before the War. Stouffer et al. (1949) coined the term “relative deprivation” to remind us how people’s subjective sense of themselves, within the context of receiving or lacking equal opportunities or wealth, can vary dramatically from their situation as described by objective measures. Through the lens of relative deprivation, being poor and feeling poor are different phenomena. Feelings of discontent felt due to inequities, thus, are not necessarily a function of those inequities as assessed by quantitative measures.
Our chapter proceeds in three main parts. In the first, we look at relative deprivation – the concept, the models, and the empirical research, especially recent research. Next, we examine cognate theories and research of those theories. Our final section centers on applications of relative deprivation. Our chapter recalls the chapter on relative deprivation that appeared in the first edition of this book (Carrillo, Corning, Dennehy, & Crosby, 2011), summarizing some of the early work, but here we emphasize work that has appeared in the last few years.
Relative Deprivation (RD)
The Concept
RD has traditionally referred to one of two things: a concept or a theoretical model (Carrillo et al., 2011). Crosby et al. (1986) argued that a concept of relative deprivation may refer to the ways individuals may or may not feel deprivation, based on subjective comparisons between the self and others. Concepts, then, are merely used heuristically to bundle together observations about related phenomena. Concepts are not empirically-tested explanations of such phenomena. It is theoretical models that provide explanations, linking antecedents and consequences. Any model of RD contains a set of propositions that may be tested empirically to then explain related phenomena. Even if no model of RD is empirically validated, the concept may be a useful one if it allows us to describe realities with parsimony (Crosby et al., 1986)
Early Models of RD
Four related but distinct models of RD dominated scholarship in the 1950s, 1960s, and early 1970s. These theories grew out of the work of Stouffer et al. (1949).
Davis (1959). James Davis (1959) believed that three factors precede the experience of RD, which we will refer to in this chapter as antecedents. These antecedents were:
Awareness that a similar other possesses X
Desire of X
Feeling deserving of X
Per Davis’s (1959) model, each of these conditions must occur for a person to experience the feeling of deprivation.
Davis (1959) hypothesized that there were four possible psychological consequences of unequal comparisons between a person, or “ego” and a referent other, or “alter.” He differentiated between ingroup comparisons, in which people compare themselves against individuals in an ingroup or similar referents, and outgroup comparisons, in which referents are of different status or rank. According to Davis’s theory, when people compare themselves to an ingroup referent, they will experience RD if they have less than the referent. But if individuals have more than the ingroup referent to whom they are comparing themselves, the person will feel gratified by relative comparison. A person who judges that an outgroup referent has more will feel relatively subordinate to that referent; if an outgroup referent is considered to be deprived, the person will feel relatively superior (Davis, 1959). Davis hypothesized that when the deprivation status of individuals and their referents are discrepant, this would result in the awareness either of “unfairness” (1959, p. 283) for ingroup comparisons, or “social distance” (p. 284) for outgroup comparisons.
Runciman (1966). William Runciman (1966) developed his theory of RD in an attempt to explain how working-class individuals came to support the Tories, a right-wing political party associated with upper-class British voters. In his model, Runciman first delineated and emphasized the distinction between personal deprivations, which he called “egoistical deprivations,” and group deprivations, which he called “fraternal deprivations.” In this chapter, we will refer to personal deprivations as individual relative deprivation (IRD) and group deprivations as group relative deprivation (GRD).
In IRD, individuals feel deprived on their own behalf; in GRD, individuals feel that their group is relatively deprived in comparison to another group. Individuals’ likelihood of experiencing IRD or GRD depends on their group involvement; if they identify strongly with their class or peer group, they are more likely to feel deprived on behalf of their group and experience GRD; on the contrary, if they have aspirations that would move them beyond their group, they are more likely to be susceptible to IRD. Runciman agreed with Davis’s (1959) RD model, but added a fourth component – feasibility. Runciman felt that RD was dependent on people’s hopes being disappointed; people do not feel deprived when a desired outcome seems either impossible or, at the other extreme, imminent and feasible (Runciman, 1966).
Gurr (1970). Ted Robert Gurr (1968a, 1968b, 1968c, 1969, 1970) was interested in RD and social violence. He proposed that RD arises out of the discrepancy between people’s “value expectations” and their “value capabilities.” Value expectations refer to those desired objects (and opportunities) to which people feel entitled. Value capabilities refer to what people think they can achieve.
Gurr’s model differed from those of other theorists, as it was dynamic. Gurr identified three possible patterns of RD: (1) decremental deprivation; (2) aspirational deprivation; and (3) prospective deprivation. Decremental deprivation refers to the ways in which people’s value expectations remain unchanged, but their value capabilities decrease. An example of this would be a person losing his or her job or taking a large pay cut but still expecting to maintain the standard of living to which the family had become accustomed. The next pattern of RD is aspirational deprivation, wherein individuals’ value capabilities remain the same, but their expectations rise. This can be illustrated by de Tocqueville’s famous observation: “Evils which are patiently endured when they seem inevitable become intolerable when once the idea of escape from them is suggested” (quoted in Davies, 1962, p. 6). Finally, in prospective deprivation, a person’s value expectations rise while, simultaneously, his or her value capabilities fall.
On the role of feasibility in RD, Gurr (1970) disagreed with Runciman (1966). Runciman’s model maintained that people would experience deprivation, when they believe that attainment of the desired is within the realm of possibility. Gurr believed the opposite – people experience RD when they believe that the desired is impossible. Early research (e.g., Coch & French, 1948) found mixed support for Gurr’s approach, but Crosby’s (1982) study of working women, which we discuss later in this chapter, validated Gurr’s position on feasibility.
Williams (1975). Robin Williams’s (1975) theory of RD was simpler than the theories of Runciman (1966) and Gurr (1970). Williams believed that RD will arise when a discrepant comparison is made between what one wants and what one has. Williams was more interested in the consequences of collective deprivation than in antecedents or in IRD. Some of these consequences include protests, revolutions, or rebellions, racial discrimination or bias, or dissatisfaction with sociopolitical and socioeconomic outcomes.
Crosby Brings Order to the Literature
By the mid-1970s, hundreds of studies had been published throughout the social sciences that used the term “relative deprivation.” The findings were chaotic in part because different meanings were attached to the same terms and in part because different researchers examined phenomena at differing levels of analysis. In some studies, researchers measured feelings or thoughts while in others researchers imputed feelings or thoughts on the basis of gross measures of behavior.
In 1976, Faye Crosby published an influential paper in Psychological Review that helped to synthesize and put order in the earlier work. Crosby first distinguished between distal and proximal antecedents of feelings of deprivation, resentment, and discontent. Some researchers like Gurr looked at distal antecedents such as the distribution of wealth in society or workplace harassment, while others, like Runciman, hypothesized proximal antecedents such as people’s thoughts and feelings.
Following most closely the work of Runciman (1966), Crosby’s model stated that five proximal antecedents must be fulfilled in order for a person to experience RD. Specifically, a person must:
See that other possesses X
Want X
Feel that one deserves X
Think it feasible to obtain X
Lack a sense of responsibility for failure to possess X.
The last factor was derived from the work of Patchen (1961) and marked a spot where Crosby parted ways from Runciman (1966).
Crosby supplemented her model by delineating what would occur if any of the hypothesized antecedents were absent (Crosby & Gonzalez-Intal, 1982). If the second precondition were not met, said Crosby and colleagues, then a person would feel “righteous indignation” rather than deprivation. If the third precondition were absent, then the feeling would be one of disappointment rather than deprivation. If the fourth precondition were not filled, the resulting state would be one of “dissatisfaction or jealousy;” and if the fifth precondition were to go unsatisfied, then “envy or dissatisfaction with himself” would result (Crosby, 1976, p. 91; Crosby & Gonzalez-Intal, 1982).
Crosby sought to test empirically her theoretical model. She undertook a large study of employed men, employed women, and housewives to document the dynamics of feelings of dissatisfaction and deprivation. At the basis of this work was the paradox of the contented female worker. Despite the fact that women were earning much less than men, research at that time showed that women expressed just as much satisfaction in their jobs as men (e.g., Burke & Weir, 1976; Deaux, 1979; Glenn et al., 1977). Indeed, women demonstrated very little subjective awareness at that time of their objective deprivation (DeLamater & Fidell, 1971; Epstein, 1971; Kresge, 1970; Widom & Burke, 1978; Young et al., 1980).
Crosby saw the construct of RD as clearly relevant to the situation of women, and at the same time, designed a study that assessed all the putative antecedents of felt deprivation as well as deprivation itself in three different populations (employed women, employed men, housewives), concerning both one’s own situation and the situation of working women. Newton, Massachusetts, was chosen as the site of the study because of its demographic mix and its reputation for accurate town data.
Participants – all aged 25 to 40 – were employed men, employed women, and housewives selected through a process of stratified random sampling. Occupations were categorized using a national rating system into high-prestige, middle prestige, and low-prestige groups. Middle-prestige workers were eliminated from the sample to ensure a clear distinction between low-prestige and high-prestige workers. Housewives were classified on the basis of their husbands’ occupations. Appointments were scheduled with 405 respondents via telephone. Participants were then interviewed in their homes by professional interviewers, for approximately one hour each. Interviewers gathered data about participants’ home, work, and family situations. Women and men were matched by the prestige level of their jobs.
Crosby’s attempt to test the various models of relative deprivation followed a simple logic. Victory would go to the model that best-predicted feelings of resentment in the different samples (i.e., employed men, employed women, housewives) across different domains (i.e., home life, work life, opinions of the situation of women). To Crosby’s disappointment, none of the models proved consistently valid. The data did not support unequivocally any single model.
Yet, a very important pattern of findings emerged unexpectedly from the data, a finding that Crosby came to call “the denial of personal discrimination” (Crosby, 1984). Objectively, the employed women in the sample were victims of discrimination. To her surprise, Crosby found that the salaries of employed women were significantly lower than those of their male peers. The difference in earnings could not be explained as a function of differential job prestige, job commitment, years on the job, hours worked each week, or background factors such as education. Everyone in the sample, and especially the employed women, knew that women were generally discriminated against. And yet the women in the sample showed no resentment, nor even much awareness of their own disadvantaged situation. It was as if each woman imagined herself to be the exception to the general rule of sex discrimination. So much for the naïve concept that people exaggerate their own misery while being blind to the misery of others.
Dozens of studies, reviewed by Crosby and Bearman (2006), have since shown that among many groups, individuals tend to underestimate the chances that they personally suffer from the inequities and discriminations that they know to affect the groups to which they belong (see Crosby & Bearman, 2006, for a review). The employed women living in Newton and studied by Crosby were not unusual in that regard. Crosby’s surprising findings were not, in fact, anomalous.
Various explanations have been offered to account for the discrepancy between deprivations felt on behalf of oneself and deprivations felt on behalf of the group (Crosby & Bearman, 2006). For example, some researchers suggested that people overestimate the extent of group deprivation rather than underestimating the extent of their own deprivation (Taylor et al., 1994). Other researchers have wondered if the apparent differences between IRD and GRD are in fact the result of imprecise or inadequate measures of either IRD, GRD, or both. Could the differences be a methodological fluke?
Data from two related studies show that the IRD-GRD discrepancy is not just a methodological flux. In the first study, Alexandra Corning (2000) developed a scale of IRD that is psychometrically robust and that allows researchers to test propositions with greater precision than was possible before. Her scale asked respondents about the extent to which they wanted some outcomes, felt deserving of the outcomes, saw others as obtaining those outcomes, and felt they personally had been encouraged to obtain the outcomes. In the 26-item Perception of Social Inequality Scale—Women’s Form (PSIS-W), Corning included questions about six domains she had identified in the literature on gender equality. They are: (a) disproportionate encounters with harassment and concern with assault, (b) differential standards of physical attractiveness, (c) differential availability of academic role models, (d) relative lack of career encouragement, (e) employment compensation based on factors unrelated to competence, and (f) differential struggle with multiple roles.
Later Corning et al. (2002) created a group, or collective, version of the PSIS—W. Exactly parallel in structure to the PSIS—W, the Perception of Social Inequality Scale—Women’s Form/Collective RD (PSIS-W/CRD) allowed for a measure of collective, or GRD. In contrast to the items of the PSIS—W, which refer specifically to women’s feelings regarding their own personal situations, the items of the PSIS—W/CRD concern the situation of women in general. Each item of the PSIS—W/CRD was created to match an item on the PSIS—W, but with the preface of each question pertaining to the average woman (relative to men) rather than the self.
Corning et al. (2002) administered both the PSIS—W and the PSIS—W/CRD to 135 undergraduate women attending a west coast university in the United States, all of whom were enrolled in a psychology course for which they received course credit for their participation in psychological research. Corning et al.’s major finding was that participants as a group expressed significantly more GRD than IRD, even though the two measures were highly inter-correlated. Their findings suggested that the oft-repeated finding that people experience more GRD than IRD was not due to measurement error.
While there was a reliable difference between experiencing GRD versus IRD, other findings in the study by Corning et al. (2002) were less dramatic. Using previously published scales, Corning et al. (2002) also assessed participants’ feelings of stress and their activism. Corning and her colleagues expected to find a strong relationship between IRD and stress, and also between GRD and activism. In fact, GRD showed a slightly higher association with one measure of stress than IRD, while IRD showed a slightly higher association than GRD on another measure of stress. However, the associations between GRD and two measures of collective action were, as predicted, stronger than the associations between IRD and the same two measures of collective action.
Thomas Pettigrew and Leaps into the 21st Century
Like many social psychologists of her generation, Crosby had been deeply influenced by the work of Thomas Pettigrew. In particular, Pettigrew’s Nebraska Symposium article on social evaluations (Pettigrew, 1967) had been one initial igniter of Crosby’s interest in relative deprivation theory. Ironically, long after Crosby moved away from her interest in the theory of relative deprivation, Pettigrew continued his work on the concept, alone and in concert with many of his colleagues and students.
Pettigrew’s championing of the concept of relative deprivation has been affirmed by empirical work that continues to demonstrate the pattern of findings of Stouffer et al. (1949). For example, with a sample of 5000 British employees, Clark and Oswald (1996) looked at actual income, “subjective” income (as indexed by the individual’s income relative to others with similar jobs), and satisfaction. They found a strong positive relationship between subjective income and satisfaction but a negative relationship between actual income and satisfaction. In fact, contrary to what economists might expect, those who had less actual income reported higher feelings of satisfaction than those with higher incomes.