Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The ageing workforce has been described as the defining social issue of the twenty-first century (Pitt-Catsouphes 2007), prompting a recent global policy orientation towards extending working life and encouraging ‘older’ workers to remain in the labour force. While there is no agreement on a definition of an ‘older’ worker, there is a broadly shared view that, in the workplace context at least, one becomes identified as ‘older’ somewhere between the chronological ages of 50 and 55 years of age (McCarthy et al. 2014). The continued labour force participation of these ‘older’ workers is now recognised as critical in facilitating economic growth (DELSA 2006; Feyrer 2007) and reducing fiscal strain on national pension and social protection systems (Heraty and McCarthy 2015). It is steadily emerging, however, that many organisations across the globe have serious reservations about employing ‘older’ workers, to the extent that negative attitudes towards older workers now appear widespread (DELSA, 2006; Posthuma and Campion 2009). As a result, the study of ageist attitudes at work has been the focus of considerable research effort in recent years. The intention of this chapter is to review relevant literature that is pertinent to our understanding of such attitudes in the workplace. We first address the nature of attitudes, and more specifically, the nature of attitudes towards ‘older’ workers, before exploring theoretical perspectives on the determinants of ageist attitudes at work; we conclude with recommendations for future research.

The Nature of Attitudes

There are numerous definitions of the term ‘attitude’: from Allport’s (1935: 810) classic conceptualisation of an attitude as ‘a mental and neural state of readiness, organised through experience, exerting a directive or dynamic influence upon the individual’s response to all objects and situations with which it is related’ to more recent classifications, such as Eagly and Chaiken’s (1993: 1) ‘psychological tendencies that are expressed by evaluating a particular entity with some degree of favour or disfavour’. This particular definition is considered the most ‘conventional’ in the contemporary attitudinal literature. Despite subtle differences across the evolution of study on the concept, most researchers in the field have emphasised the evaluative aspects of an attitude (see Table 16.1) in that an attitude involves the expression of an evaluative judgement about an object. Indeed, as Maio and Haddock (2010: 4) state, ‘most attitude theorists would argue that evaluation is the predominant aspect of the attitude concept’.

Table 16.1 Evaluative nature of attitudes
Full size table

Furthermore, Eagly and Chaiken (1993, 1998) suggest that when attitudes are conceptualised as evaluative judgements, they can vary in two central ways: first, attitudes differ in terms of valence, in other words, they can be positive or negative and second, attitudes differ in strength, consequently an individual can feel less or more strongly about an object than others. ‘People are sensitive to covariations that they observe between the presence of a given object and the presence of related positive and negative cues’ (Fazio et al. 2004: 294).

Certainly, an underlying assumption across the literature is that an attitude can be reduced to the net difference between the positive and negative stimulations associated with an object (Allport 1935; Thurstone 1931). Likewise, people’s feelings about an object can vary ‘anywhere between two endpoints: maximally positive (and minimally negative) to maximally negative (and minimally positive)’ (Cacioppo and Bernston 2004: 401). Shook et al. (2007) found that extreme attitudes are more influential and are thus given more weight than mild attitudes. As such, strong attitudes are considered consequential; these attitudes can be resistant to change, persistent over time and predictive of behaviour (Eagly and Chaiken 1993; Pomerantz et al. 1995). While there has been considerable debate about the attitude–behaviour relationship, it is commonly acknowledged that an attitude is a predisposition to behave in a particular way (Fazio 1986, 1995; Procter 2003), and, as Schuman et al. (1997: 6) point out, ‘if attitudes and behaviours existed in entirely different spheres, learning about attitudes would be of little practical value, whatever their interest from the standpoint of intellectual understanding. But careful reviews of a wide range of past studies, as well as specific experimental research, make it clear that this is not the case. Attitudes and relevant behaviour at the individual level are usually correlated to some extent’.

Moreover, attitudes are generally considered to comprise cognitive, affective and behavioural components. The cognitive domain refers to the beliefs about the probability that an object is associated with a given attribute (Fishbein and Azjen 1975). Affect encompasses an overall emotional feeling concerning an object (Berkowitz 2000), and behaviours are generally defined as the intended actions of an individual (Fazio 1986, 1995). For example, Triandis (1971: 2) explains this ‘tripartite view’ that an ‘attitude is an idea (the cognitive component) charged with emotion (the affective component) which predisposes a class of actions (the behavioural component) to a particular class of social situations’. More recently, Eagly and Chaiken (1993, 1998) extend this view to propose that this tripartition best represents the types of responses that allow researchers to identify attitudes because individuals’ attitudes are shaped based on their cognitive, affective and behavioural responses to an object. Individuals express their attitudes by means of holding certain beliefs about an object (cognitive domain), feeling a certain way about an object (affective domain) and intending to behave in a certain way (behavioural domain). It is suggested, therefore, that an attitude is actually an ‘evaluative summary’ of the information derived from these bases (Fabrigar et al. 2005; Zanna and Rempel 1988), rather than simply ‘consisting’ of cognitive, affective and behavioural elements. While these components are recognised as being empirically distinct, they are often ‘directionally consistent’, for example, positive beliefs about a social group can be associated with positive feelings about the group, and in turn, linked with positive behavioural intentions towards this group (Breckler 1984; Zanna and Rempel 1988). We now turn to the nature of attitudes specifically with respect to age.

The Nature of Ageist Attitudes

We have seen that evaluation is the predominant tenet of an attitude, and Nelson (2005) proposes that people automatically evaluate other people along three particular dimensions: race, gender and age. As people categorise others along these dimensions, often they develop attitudes. Attitude research has suggested that age is one of the first characteristics we notice about other people (Kite et al. 1991; Fiske 1998; Cuddy and Fiske 2002). Research has also demonstrated that attitudes are more negative towards older than younger adults (Palmore 1999; Kite et al. 2005). That is, that ageism exists. Ageism has been described as the third great ‘ism’, following racism and sexism, but has received much less attention across the bias literature (Blancato and Ponder 2015; Shore et al. 2009). Duncan (2001) notes that there has been less focus on preventing discrimination from ageist attitudes than on discrimination from racist and sexist attitudes. He believes that this is rather ‘ironic’ (Duncan 2001: 26), especially considering that ageist attitudes have the potential to affect everyone as they get older, not just members of one particular race, sex or other demographic grouping (Achenbaum 2015). Robbins (2015) suggests that the lack of attention to ageism in comparison to racism and sexism is because ageism is often ignored or, indeed, even accepted in modern society, and this is, of course, ‘the pernicious problem of ageism’ (p. 6).

Butler’s (1969) seminal work coined the term ageism to refer to evaluative judgements towards a person or persons simply due to their advanced age. Here, he states that ageism comprises three interrelated aspects: prejudicial attitudes towards older persons, old age and the ageing process; discriminatory practices against older people; and institutional practices and policies that perpetuate stereotypes about older adults (Butler 1969, 1980). In their review of the concept of ageism, however, Iversen et al. (2009: 8) state that ‘it is problematic that many researchers are still using this definition as the basis of empirical studies’. First, Butler’s (1969, 1980) definition implies that only ‘older’ people experience ageism. ‘Younger’ people, too, can be subjected to evaluative judgements and negative attitudes based on their perceived ‘youth’. Second, Butler’s (1969, 1980) definition does not use the classic cognitive–affective–behavioural tripartition because the cognitive component is not included. Instead, it focuses on prejudicial attitudes (affective domain), discriminatory practices and institutional practices (behavioural domain). Although ageism has been defined using the tripartite view by some theorists, Iversen et al. (2009) point out that many of these definitions do not explicate the tripartite structure. The tripartite structure of ageism is perhaps more comprehensively depicted by Kite and Wagner (2004), who identify that ageist attitudes comprise an amalgam of stereotypes (cognitive domain), prejudice (affective domain) and discriminatory intentions (behavioural domain). In addition to the tripartite view, it has also been acknowledged that ageist attitudes comprise positive and negative components, and can be both explicit (conscious) and implicit (unconscious) (Cuddy and Fiske 2002; Levy and Banaji 2002; Malinen and Johnston 2013). Palmore (1999) and Solem (2007) further make a distinction between ageism on an institutional level (as referred to by Butler 1980) and ageism on an individual level. Institutional ageism refers to an overall climate of ageism among members of organisations and institutions, including organisational policies and practices which serve to negatively affect individuals with respect to their age. Individual ageism, on the other hand, represents the ageist attitudes held by individuals in society. Table 16.2 provides the range of definitions of ageism that have been proposed over the past 50 years.

Table 16.2 Definitions of ageism
Full size table

Seeking to offset the inconsistency among theorists on the complexity of ageist attitudes, and with the purpose of providing clarity and accuracy on the concept, Iversen et al. (2009: 15) offer this definition of the term: ‘Ageism is defined as negative or positive stereotypes, prejudice and/or discrimination against (or to the advantage of) people on the basis of their chronological age or on the basis of a perception of them as being “old” or “elderly”. Ageism can be implicit or explicit and can be expressed on a micro-, meso-, or macro-level’.

We argue that this particular view of ageism is, perhaps, the most comprehensive, and indeed, the most practical definition in the literature to date for a number of reasons. First, the traditional social psychological tripartite view of attitudes is encompassed in the cognitive–affective–behavioural domain; second, it recognises that attitudes can be both negative and positive; third, the conscious (explicit) and unconscious (implicit) elements of ageism are acknowledged; additionally, this definition makes specific reference to the perceptual nature of how individuals categorise people by age. Most significantly, the operation of ageist attitudes at the individual (micro) level, the social network (meso) level and the institutional and cultural (macro) level are particularly useful in studying ageist attitudes at work because ageist attitudes can exist among individuals, in groups and teams, among professions and within professional networks, and can permeate industry, organisational and societal cultures. We do have one criticism, however, where we believe that this definition is limited by its focus on ‘old’ or ‘elderly’. Where ageism has also been found to exist against ‘younger’ and indeed ‘mid-life’ adults, we advocate for the removal of the phrase ‘or on the basis of a perception of them as being “old” or “elderly”’ in this definition, instead using ‘or on the basis of a perception of them as being “old” or “elderly”, “young” or “mid-life”’.

Ageist Attitudes in the Workplace

In the work environment, research shows that negative attitudes about older employees exist (cf. Tuckman and Lorge 1952; Kirchner and Dunnette 1954; Bird and Fischer 1986; Hassell and Perrewe 1995; Chiu et al. 2001; Kite et al. 2005; Loretto and White 2006; Posthuma and Campion 2009; Ng and Feldman 2012) to the extent that we can say ageism exists in many organisations. Where ageist attitudes comprise stereotypical, prejudicial and behavioural components, it appears that stereotypes of ‘older’ workers have been the central focus of previous research on age-related attitudes in the workplace.

Stereotypes

Workplace age stereotypes are beliefs and expectations about workers based on their age (Hamilton and Sherman 1994; Posthuma and Campion 2009). The term ‘stereotype’ is attributed to the Parisian printer, Didot, who first used the word in 1798 to describe a printing process that created reproductions using moulds (Ashmore and Del Boca 1981). This expression has evolved into a metaphor for mental reproductions of reality (Nelson 2004). As such, generalised beliefs individuals have about members of particular groups in society are usually labelled as stereotypes. As Lippman (1922: 81) stated, ‘for the most part we do not first see, and then define; we define first and then see. In the great blooming, buzzing confusion of the outer world we pick out what our culture has already defined for us, and we tend to perceive that which we have picked out in the form stereotyped for us by our culture’.

As we see here, Lippman (1922) likened stereotypes to ‘pictures in the head’ which come to mind quickly when we think about groups or members of groups in society, believing that these ‘pictures’ help us make sense of our world and our reality. Zanna and Olson (1994) emphasise the general agreement that stereotypes are beliefs; more specifically, a stereotype is a cognitive construct (Fishbein and Azjen 1975). Stereotypes, however, are recognised as a ‘relatively simplex cognition, especially of a social group’ (Krech et al. 1962: 67), which is exaggerated. Age-based stereotypes generally regard old age as a period of poor health, loneliness, resistance to change as well as declining physical and mental abilities. Bytheway (1995) found that when people are defined as old, they are often categorised as senile, rigid, old-fashioned and inferior. Palmore (1999) identified nine major stereotyped characteristics associated with ‘older’ people as illness, impotency, ugliness, mental decline, mental illness, uselessness, isolation, poverty and depression. Although mostly negative, positive stereotypes about ‘older’ people also exist, where they are labelled as, for example, kind, wise, dependable, affluent and powerful (Palmore 1999). Further, Cuddy et al. (2005) proposed their ‘Stereotype Content Model’ to highlight that ‘older’ people are often stereotyped along two dimensions: competence and warmth. The warmth dimension is characterised by positive traits such as friendliness and honesty and by negative traits such as coldness and untrustworthiness. The competence dimension, then, is characterised by positive traits such as assertiveness and intelligence and by negative traits such as inefficiency, indecisiveness and laziness (Cuddy et al. 2005). It is now understood, therefore, that stereotypical beliefs about groups are usually not entirely negative (or positive), but contain a mixture of positive and negative elements (Fiske et al. 2002).

Stereotypes of ‘older workers’, however, attribute mostly negative work-related characteristics to this group (Posthuma and Campion 2009). For example, ‘older’ workers are seen as being resistant to change and having a lower physical and performance capacity (Rosen and Jerdee 1977; Gordon and Arvey 2004), being less motivated (Craft et al. 1979), less able to handle criticism (Tuckman and Lorge 1952), making fewer contributions to the organisation with a lower potential for development (Perry and Varney 1978), unable to work in teams (Lyon and Pollard 1997) and less economically beneficial (Finkelstein et al. 2000) than younger workers. Hayward et al. (1997) found beliefs that ‘older’ workers are harder to train than younger workers as well as being too ‘cautious’ at work. Some positive stereotypes about older workers have also been found in the literature (e.g. Kluge and Krings 2008). For example, ‘older’ workers are perceived to be more reliable (Hayward et al. 1997), more conscientious (Warr and Pennington 1993), with lower rates of absenteeism (Broadbridge 2001; Hedge et al. 2006) and better people skills (AARP 2000) than ‘younger’ workers.

Interestingly, discussions about stereotypes of ‘older’ workers often draw comparisons with ‘younger’ workers, but relatively little is known about stereotypes of ‘younger’ workers, and this is a limitation of our understanding of ageist attitudes. Ageism is not limited to ‘older’ workers, as it also affects ‘younger’ workers who may be stereotyped as being too young, as well as lacking the necessary skills and experience for particular roles (Snape and Redman 2003). For a discussion of emerging research on age stereotypes as they relate to ‘younger’ workers, please see the chapter by Nadler et al. in this volume.

While stereotypes have been the most widely researched component of ageist attitudes at work, it is now accepted that attitudes about groups may not be derived solely from stereotypical beliefs (Haddock and Zanna 1998). We have already discussed how ageist attitudes comprise not only stereotypical beliefs but also prejudicial feelings and behavioural predispositions. Accordingly, the following section places an emphasis on affective (prejudice) and behavioural (discriminatory predispositions) attitudes towards ‘older’ workers.

Affective and Behavioural Attitudes

Affective attitudes encompass individuals’ overall emotional feelings concerning an object (Berkowitz 2000; Fiske et al. 2002), while behavioural attitudes are generally defined as the intended actions of an individual towards an object (Fazio 1986, 1995). With respect to ageist attitudes, affect represents prejudicial feelings, while behaviour represents discriminatory predispositions, towards an individual based on their age or a perception of their age (Kite and Wagner 2004; Iversen et al. 2009). Much of the research on ageism, however, has been criticised for assessing only the cognitive components of ageist attitudes, namely, stereotypes (Fraboni et al. 1990; Rupp et al. 2005; Finkelstein and Farrell 2007), where the affective and behavioural components of ageist attitudes are often neglected.

Finkelstein and Farrell (2007: 76) note that the affective component of attitudes towards ‘older workers’ appears to be the ‘least consistently conceptualised and measured in the bias literature’, where efforts to separate affect from cognition and behaviour are only now beginning to be discussed. We add that research on affective attitudes towards ‘younger’ or ‘mid-life’ workers is also lacking. Discrimination itself, in terms of actual behaviour towards ‘older’ workers, has provided some focus in the literature because it is acknowledged that it exists. First, some workers perceive that they are discriminated against because they are ‘older’, leading to an increase in age lawsuits across the globe. It is also now recognised that the categorisation of workers as ‘older’ has a negative impact on the employability of this group of workers (McCarthy et al. 2014). Discrimination against ‘older’ workers has been established in employment-related outcomes such as selection, participation in training and opportunities for promotion. Raza and Carpenter (1987) found that age was negatively related to hire-ability, while Kanfer et al. (2001) demonstrated that ‘younger’ people reported a greater likelihood of becoming re-employed than ‘older’ people. There is also evidence to suggest that ‘older’ workers are less likely to gain access to training and development opportunities at work, and may also be more likely to be passed over for promotion in favour of younger employees (Palmore 1990). Further, ‘older workers’ are considered more likely to be selected for redundancy than younger age groups (Snape and Redman 2003), where supervisors are not in favour of employees working up until retirement age (Henkens 2000). Experiences of age discrimination have been found to be increasingly common for ‘younger’ employees, with under-25s being twice as likely to experience age-based discrimination compared to individuals in other age categories (Snowdon 2012).

However, there appears to be a lack of systematic empirical evidence on assessing people’s actual discriminatory predispositions, in other words, their intentions to (or not to) discriminate (Chiu et al. 2001). Many studies concerning age discrimination are limited to discussing age stereotypes. Yet, we know that these attitude constructs are conceptually and empirically distinct. Stereotypes, of course, are often seen as a precursor to discriminatory behaviours (Dovidio et al. 1996), but research suggests that attitudes based on emotion can be stronger, or at least more stable, than attitudes based on beliefs (cf. Edwards 1990; Edwards and Von Hippel 1995; Giner-Sorolla 2001). Moreover, behaviour is thought to be affectively driven (Esses and Dovidio 2002). This reinforces the need to address the affective and behavioural dimensions of ageist attitudes, where these components are seen as ‘important to measurement in future research on age bias – especially workplace bias’ (Rupp et al. 2005: 356).

Measuring Ageist Attitudes

Much of the literature on attitudes is specifically concerned with the development of instruments designed to measure attitudes, and while there has been much debate on the measurement of the attitude construct, Thurstone (1928: 530) wrote: ‘It will be conceded at the outset that an attitude is a complex affair which cannot be wholly described by any single numerical index. Nevertheless, we do not hesitate to say that we measure a table…Just in the same sense we shall say here that we are measuring attitudes. We shall state or imply by the context the aspect of people’s attitudes that we are measuring. The point is that it is just as legitimate to say that we are measuring attitudes as it is to say that we are measuring tables or men’.

Anderson (1981) notes that information about individuals’ attitudes can be measured in two fundamental ways: either through observational methods (the indirect approach) or through self-report methods (the direct approach). Research involving indirect measures of attitudes typically takes place within a laboratory setting, using techniques such as word association tasks (Fazio et al. 1995), the implicit association test (Greenwald et al. 1998) and recall of stimuli (Schneider 2005); however, participants are usually unaware of what is being measured. While indirect methods have been successful in predicting unconscious bias (Dovidio et al. 1997), these types of methods obviously have huge implications for feasibility in terms of resources and practicality, as well as ethical considerations for the participants.

The direct approach, on the other hand, assesses attitudes using self-report methods, where participants are normally aware of the types of attitudes that are being measured, and have control over the measurement outcome (DeHouwer and Moors 2010). These self-report methods usually comprise a series of questions or statements about an attitudinal object, where participants are asked to give an evaluation of this object, which is then recorded (Anderson 1981; DeHouwer and Moors 2010). Problems inherent in the direct approach relate to the potential misinformation provided by the participant because of a desire to respond in a socially acceptable manner. However, Karpinski and Hilton (2001) and Dovidio et al. (2002) demonstrated that direct methods for measuring individual attitudes were better in predicting individual choices than indirect methods, while Griffiths (1999) stated that direct approaches do not restrict research paradigms, nor do they require the types of resources associated with laboratory research, and are therefore suitable in organisational research.

Although researchers over the past few decades have developed various, and mostly unidimensional, scales measuring attitudes towards older people in general (cf. Rosencrantz and McNevin 1969; Fraboni et al. 1990), few of them have constructed scales specifically measuring attitudes towards ‘older’, or indeed ‘younger’, workers directed at organisational decision-makers. Those that have, place an emphasis on stereotypical attitudes of older workers. A number of published measures on stereotypes of ‘older’ workers have been advanced based on the work of Tuckman and Lorge (1952), Kirchner and Dunnette (1954), Bird and Fischer (1986), Hassell and Perrewe (1995) and Chiu et al. (2001). Measures concerning the affective dimension of attitudes towards ‘older’ people exist (Fraboni et al. 1990; Rupp et al. 2005) but, to the best of our knowledge, no measures exist to assess the prejudicial attitudes towards ‘older’ (or ‘younger’) workers more specifically. Finally, measures concerning the behavioural dimensions of attitudes towards ‘older’ workers have been advanced by Chiu et al. (2001), but these are limited to single-item measures. We call for the development of instruments which seek to measure the cognitive, affective and behavioural dimensions of ageist attitudes within the decision-making context at work.

Influential Factors

Ageist attitudes towards ‘older’ workers in organisations have been found to exist particularly within an employment-related decision-making context (cf. Hassell and Perrewe 1995; Shore et al. 2003; Posthuma and Campion 2009). Several characteristics of this decision-making context appear to have some influence on the positive or negative nature of ageist attitudes at work, including the individual characteristics of raters (the decision-makers) themselves and the characteristics of the organisation within which this decision-making occurs.

Individual Factors

As noted earlier in this chapter, attitudes can vary in terms of valence and strength, such that an individual can feel more or less favourable towards an object than others (Eagly and Chaiken 1993, 1998). A number of studies have examined whether the age of the rater (the individual evaluating a group) influences their attitudes towards ‘older people’ or towards ‘older workers’. Rothbaum (1983) and Chasteen et al. (2002), for example, found that older raters held more positive attitudes towards older people, while Bird and Fischer (1986), Hassell and Perrewe (1995) and Chiu et al. (2001) found that older employers held more positive stereotypes towards ‘older’ workers than did ‘younger’ employers. These findings are often attributed to Social Identity Theory (Tajfel and Turner 1979; Ashforth and Mael 1989), where individuals can derive a sense of belongingness and worth from their membership in social groups, and so, they are more likely to hold more favourable attitudes towards members of their own social group. ‘Older’ employers in the findings noted above, may therefore derive a sense of identity with ‘older’ workers, and hence be inclined to evaluate them in a more favourable light. Findings from research, however, do not always indicate that ‘older’ raters will favour ‘older’ workers. For example, Finkelstein and Burke (1998) found that older raters demonstrated no preference for ‘older workers’ over younger workers. Kite et al. (2005), on the other hand, found that older individuals stereotype ‘younger workers’ as more competent than ‘older workers’. The evidence in this area of research is certainly mixed, but it nonetheless demonstrates, that in some cases, attitudes towards ‘older workers’ can be influenced by the age of the person evaluating these ‘older workers’.

There is also some evidence to suggest that gender considerations are an important factor, where female employers have been found to hold more positive beliefs about ‘older workers’ than male employers (Kogan and Shelton 1962; Rosen and Jerdee 1976; Connor et al. 1978; Kalavar 2001; Rupp et al. 2005). The only explanation for this finding offered in the literature thus far is that it may be partly due to higher levels of expressiveness in personality on the part of women (Deaux 1985; Rupp et al. 2005). As such, women are believed to be less critical and more caring than men. The possible gender effect on ageist attitudes therefore warrants further attention (Rupp et al. 2005) as it appears, in some cases, to be significant, but less understood.

Additionally, two of the most widely cited studies on attitudes towards ‘older workers’ by Kirchner and Dunnette (1954) and Bird and Fischer (1986) found that supervisors held more negative stereotypes of ‘older workers’ than did rank-and-file employees. This evidence was later supported by both Hassell and Perrewe (1995) and Chiu et al. (2001) in their investigations. The position or status of a rater within an organisation is therefore purported to influence the nature of attitudes towards ‘older workers’ from these findings. Both Hassell and Perrewe (1995) and Chiu et al. (2001) implied aspects of Social Identity Theory (Tajfel and Turner 1979) when they offered an explanation for the influence of rater position. Hassell and Perrewe (1995: 466), for example, offered that, ‘Although speculative, because supervisors may be “older” themselves, they psychologically may deny membership in that category to protect their work identity and status. Older supervisors may perceive themselves to be contributing and valued members of the organisation, thus, they may not want to be viewed as an “older” employee. In order to separate themselves from the older workers, however, they may negatively stereotype them’.

Organisational Factors

Perry and Parlamis (2005) argue that the influence of organisational factors on age bias in the workplace have been relatively ignored across the literature, being far less understood than individual influences. Of the available evidence, Lucas (1995) found that negative stereotypes of ‘older workers’ were more prevalent among employers in smaller organisations than in larger organisations within the hospitality industry. She attributed this finding to less sophisticated employment policies, and more specifically, less sophisticated equality policies, in smaller firms (Lucas 1995). Her study illustrates three important points. First, although Chiu et al. (2001) found no significant relationship between firm size and attitudes towards ‘older workers’, Lucas’ (1995) research indicates that the size of an organisation may be influential in explaining attitudes towards ‘older workers’. Second, there is a possibility of an industry effect on attitudes towards ‘older workers’. Both Chiu et al. (2001) and DeMicco (1989) argue that employers in service-type industries may have more negative attitudes towards ‘older workers’ than other industries. They reason that industries dominated by frequent role contact with customers, particularly a younger demographic of customers, often favour younger over older employees. Adler and Hilber (2009) recently asserted that ‘older workers’ are under-represented in service sectors, adding some support to Chiu et al.’s (2001) and DeMicco’s (1989) views. Finally, Lucas (1995) proposes that the presence of an equality policy in an organisation may influence attitudes towards ‘older workers’. Chiu et al. (2001: 636) argue that such policies are likely to ‘cultivate a more tolerant atmosphere toward older workers in the organisation by raising awareness and countering stereotypical beliefs’. They found that the presence of an equality policy in an organisation resulted in more favourable attitudes towards ‘older workers’ (Chiu et al. 2001).

There is also some evidence to suggest that the age demographics of an organisation have an influence on attitudes towards ‘older workers’. Relational Demography (Tsui et al. 1995) posits that the demographic composition of an organisation or workgroup influences individuals’ attitudes at work. Here, it is suggested that similarity to referent others in an organisation results in favourable outcomes, much like Social Identity Theory (Tajfel and Turner 1979; Ashforth and Mael 1989), because it is based around the similarity–attraction paradigm (Byrne 1971; Riordan and Shore 1997). Individuals who are similar to one another are more likely to treat each other in a favourable manner, while dissimilarity can lead to negative treatment. This could be linked with McCain, O’Reilly and Pfeffer’s (1983) earlier research to indicate that being ‘older’ in a work group dominated by younger people results in a greater tendency for the ‘older workers’ to leave the organisation.

Relational Demography also suggests that the frequency of interactions with groups in the organisation can influence attitudes towards these groups (Tsui et al. 1995). For example, both Hassell and Perrewe (1995) and Chiu et al. (2001) found some evidence to suggest that frequency of interaction with ‘older workers’ has the potential to reduce negative stereotypes. This appears to be related to Butler’s (1969) and Falkenberg’s (1990) argument that greater interaction with social groups reduces the likelihood in the formation of bias. Remery et al. (2003), however, found that organisations with more ‘older workers’ actually had more negative beliefs about them, which as Finkelstein and Farrell (2007) state, remains without explanation.

Within the broader attitudinal literature, research on the formation of attitudes has received relatively little attention. Eagly and Chaiken (1993: 681) state that this ‘lack of attention to the development issue of how attitudes are formed and become strong… (is a) serious omission and limitation’ of the literature. More recently, however, Kite and Wagner (2004) have suggested that Social Role Theory (Eagly 1987, 1997) might be potentially useful in explaining the development of attitudes towards ‘older workers’. Social Role Theory proposes that viewing people in various social roles provides an important basis for attitudes and beliefs about social groups (Eagly 1987, 1997; Eagly et al. 2000).

Social Role Theory (Eagly 1987, 1997) proposes that our attitudes about social groups can be derived from observing, and interacting with, people in various social roles. From these interactions with group members and observations about their behaviour, we develop expectations about how all members of these groups behave. As Kite et al. (2005: 243) explain, ‘because we observe the role-driven behaviour, which may or may not reflect the real attributes of the person being observed, perceivers come to associate characteristics of these roles with the individuals who occupy them’.

Arguably, one important social role in an organisation is the role of an ‘older’ worker. As we have already seen from the discussions earlier in this chapter, the ways in which others ‘categorise’ workers as ‘older workers’ is considered especially important in determining their attitudes and behaviours towards ‘older workers’ (Sterns and Doverspike 1989; Cleveland and Shore 1992; Pitt-Catsouphes et al. 2010). This fits with the Role Theory perspective. In the search for explanations that account for individual differences in attitudes towards ‘older workers’, (or indeed, ‘younger’ workers), Role Theory may offer greater explanatory power than previous stereotype accounts (cf. Rosen and Jerdee 1977; Posthuma and Campion 2009) because Role Theory outlines how we come to derive these stereotypes in the first place; stereotypes about groups are formed from our observations and expectations of group behaviour. Role Theory, essentially, takes a step back, offering a wider lens from which to view individual differences in attitudes towards ‘older workers’ than Social Identity Theory (Tajfel and Turner 1979) and Relational Demography (Tsui et al. 1995) outlined earlier. Role Theory posits that individual differences in attitudes are less about similarity–attraction to a group, and more about how different individuals’ view the role of an ‘older worker’ that lead to their individual attitudes towards ‘older workers’.

Discussion and Future Directions

Our intention in this chapter was to provide a comprehensive review of the literature in order to advance an understanding of ageist attitudes in the workplace. Ageist attitudes can have potentially serious consequences for older workers, as they may serve to offer limited opportunities to people who are perceived as ‘older’ at work. Particularly, negative attitudes may ‘affect the judgements and actions of organisational decision makers’ (Hedge et al. 2006: 46). Understanding the nature of ageist attitudes is therefore hugely important at a time when the workforce is not only rapidly ageing but also becoming increasingly age-diverse. Despite recent and burgeoning research, there is a lack of systematic empirical evidence of the age-related attitudes that exist not only towards ‘older’ workers but also towards ‘younger’ or ‘mid-life’ workers. Poor operationalisation of the tripartite structure of ageist attitudes across research studies is evident, as are theoretical advancements on how age-related attitudes are created and maintained within organisations. As such, and based on our discussion throughout this chapter, we have four critical, and specific, directions for future research in this area: (1) Researchers, practitioners and policymakers concerned with ageist attitudes at work need to begin to focus on how ageist attitudes affect not only those perceived to be ‘older’ but also those perceived to be ‘younger’ or ‘middle-aged’ at work; (2) Further, those involved in examining ageist attitudes in the workplace need to not only account for the role that age stereotypes have to play in workplace decision making but also that of emotion (affect) and behavioural predispositions as they relate to age; (3) We call for the development of valid and reliable measures of cognitive, affective and behavioural age-related attitudes for use in the decision-making context at work; and (4) We advocate for future work in this area to be underpinned by sound theoretical foundations, which aim to advance our theoretical knowledge about how ageist attitudes are actually formed in organisations. It is only when we unearth a real, comprehensive understanding about how ageist attitudes are formed, and what forms these attitudes take within organisations, that we can begin to advance real solutions and interventions to reduce the incidence of ageist attitudes within the workplace.