Measuring positive attitudes toward persons with dementia: A validation of the Allophilia scale

Abstract

Efforts to combat ageism typically focus on negative attitudes toward members of an out-group. Changing attitudes also requires assessment and enhancement of positive attitudes. This study examined the psychometric properties of Allophilia scale when used to measure college students' positive attitudes toward persons with dementia. Data collected from 465 students were analyzed using confirmatory factor analysis and multigroup analysis by gender to assess the validity and reliability of the Allophilia scale. Results showed that the Allophilia scale is valid both for male and female students. The Allophilia scale is a valid assessment tool for measuring positive attitudes toward persons with dementia. Use of this scale will contribute to our understanding of attitudes toward persons with dementia and has implications for the design of interventions to facilitate positive attitudes toward members of this out-group.

Keywords

ageism Alzheimer disease factor analysis geriatrics psychometrics students' attitudes

Introduction

The past decade has witnessed an increase in research that explores young adults' attitudes toward older adults in general (e.g. Breytspraak, Arnold, & Hogan, 2008) and toward persons with dementia (PWD) in particular (e.g. Chung, 2009; Kaf, Barboa, Fisher, & Snavely, 2011; Yamashita, Kinney, & Lokon, 2013). The interest in the latter might stem from several factors. First of all, demographics alone implicate dementia as a significant health concern for older adults. For example, Plassman et al. (2007) estimated the prevalence of all dementias among adults age 71 and over to be 13.9% (3.4 million); the prevalence was 5.0% for participants between the ages of 71 and 79, and 37.4% for those aged 90 and older. They estimated the prevalence of Alzheimer's disease among adults age 71 and older to be 9.7% (2.4 million). Similarly, Brookmeyer et al. (2011) reported that the cause of approximately 70% of dementias in the United States is Alzheimer's disease, and this percentage increases to almost 80% among those who are 90 and older. Obviously, dementias will continue to challenge older adults and their informal and formal networks in upcoming decades.

Additionally, growing research on attitudes toward PWD may stem from the implications of the widely known concept of ageism, and the long-standing fundamental assumption rooted in psychology that stereotypes exert an influence on social behavior (Jones, 1998). Broadly defined, ageism is systematically stereotyping and discriminating against people because of their age. One consequence of ageism is that it allows younger people to distance themselves socially and emotionally from older adults to the extent that they “subtly cease to identify with their elders as human beings” (Butler, 1975, p. 35). Although a broad definition of ageism includes age-based prejudice toward people at any age (Nelson, 2009) and Butler (1980) distinguished malignant ageism (i.e. ageism that can have a harmful effect; for example, the stereotype that old people are worthless) from benign ageism (i.e. a subtle form of ageism based on one's fear and anxiety of growing old; for example, birthday cards that mock rather than celebrate advancing age), the most frequent target of malignant ageism is older adults (Nelson, 2004).

In an extreme example of stereotyping that many gerontologists find troubling, Behuniak (2011) argues that both the biomedicalization of dementia and the social construction of PWD as zombies create fear about the disease/individuals with the disease and marginalizes those with the disease. This fear translates into “dehumanization based on disgust and terror” (p. 70). This is an extreme example of what Kitwood (1990) referred to as malignant social psychology and Sabat (2003) used as the foundation for his theory of malignant social positioning.

Strategies to address negative attitudes toward older adults have been actively discussed for several decades. For example, based on the assumption that negative attitudes derive from a lack of knowledge, Palmore (1999) argued that even short-term gerontology education can reduce prejudice against older adults. Not surprisingly, the growing prevalence of dementia and the negative consequences of stereotyping PWD have resulted in an increase in interventions to improve college students' attitudes toward PWD (Chung, 2009; Kaf et al., 2011; Lokon, Kinney, & Kunkel, 2012). Although Blieszner and Artale (2001) indicated that early evidence of the efficacy of these interventions was mixed, positive effects of educational interventions (i.e. gerontology courses, service learning/internship programs) on college students' attitudes toward older adults have been reported (Allan & Johnson, 2008; Lokon et al., 2012; Palmore, 1998; Yamashita et al., 2013).

Many of these interventions are premised on the belief that, if successful, they will foster more positive attitudes among younger adults, who will increasingly encounter PWD in their personal and professional lives. A second premise is that more positive attitudes about older adults in general, and those with dementia in particular, might encourage young adults to pursue careers in a variety of health-related and social service occupations and contribute to the well-trained workforce that will be required to meet older adults' needs in upcoming years. Given the potential impact of these interventions for young adults' personal and professional lives, it is surprising that few researchers have pursued suggestions that young women and men may have different attitudes toward older adults (Kite, Stockdale, Whitley, & Johnson, 2005).

Despite the rich tradition of research on attitudes toward older adults, several long-established and frequently used scales have serious limitations. Given the two most widely used scales were developed in the 1960s, but had not been evaluated in the 21st century, Iwasaki and Jones (2008) examined the psychometric properties of the Kogan Attitudes Toward Old People scale (OPS; Kogan, 1961) and Polizzi and Millikin's (2002) revised version of the Aging Semantic Differential (ASD; Rosencranz & McNevin, 1969). They concluded that the OPS required major refinements and that, although the ASD showed more promise, its factor structure was unclear. Similarly, Cowan, Fitzpatrick, Roberts, and White (2004) conducted an extensive examination of Palmore's (1998) Facts on Aging quizzes, identifying “inadequacies, confusion, contradictions, and a need for further research” (p. 237). Other tools that have been used to measure college students' attitudes toward aging and older adults, such as the Student Assisted Independent Living Questionnaire (Pillemer & Schultz, 2002) and the Fraboni scale of Ageism (Fraboni, Saltstone, & Hughes, 1990), either are infrequently used and/or lack extensive psychometric work. Thus, there is need for a psychometrically sound scale that can be used to measure individuals' attitudes toward older adults in general and those with dementia in particular.

Since its emergence in the social science literature, ageism has typically been conceptualized in terms of “in-group” and “out-group” relations and “othering” (e.g. Bytheway, 2005; Jönson, 2013). Bodner (2009) presents an overview of social identity theory (Tajfel, 1981) and terror management theory (Greenberg, Solomon, & Pyszczynski, 1997; Solomon, Greenberg, & Pyszczynski, 2004) to explain ageism among older and younger adults. Social identity theory proposes that young adults evaluate members of their own age more favorably than older adults, from whom they distance themselves and with whom they do not identify. Terror management theory maintains that young adults are threatened by an unconscious awareness of their own death and, to protect themselves from this death anxiety, they distance themselves from and hold negative attitudes about older adults in an effort to preserve their own self-esteem (Martens, Goldenberg, & Greenberg, 2005; Martens, Greenberg, Schimel, & Landau, 2004). Terror management theory has recently been used to conceptualize and develop a psychometrically sound scale to measure attitudes toward dementia (O'Connor & McFadden, 2010).

In a novel conceptualization of out-group relations, Pittinsky, Ratcliff, and Maruskin (2008) and Pittinsky, Rosenthal, and Montoya (2011a, 2011b) maintain that the quest to reduce negative attitudes is necessary, but not sufficient, to promote positive out-group relations. Because positive attitudes foster proactive and prosocial behaviors (Pittinsky, 2009), researchers should seek to understand, and interventions should promote, positive attitudes toward members of out-groups (Pittinsky et al., 2011a). Pittinsky et al. (2011a, p. 42) refer to positive attitudes toward members of an out-group as “allophilia,” which is “derived from the Greek for ‘like or love for the other’.” Citing evidence that positive and negative attitudes are not binary, Pittinsky et al. (2011a) developed the Allophilia scale to facilitate research on positive attitudes toward members of out-groups. The Allophilia scale is a reliable and valid scale that has been used to measure five aspects of positive attitudes among members of different out-groups: affection, comfort, kinship, engagement, and enthusiasm (Pittinsky & Montoya, 2009; Pittinsky et al., 2011a).

The purpose of this research was to examine the psychometric properties of Pittinsky et al.'s (2008, 2011a, 2011b) Allophilia scale in the context of female and male college students' attitudes toward people with dementia. We examined the attitudes of females and males separately based on evidence that gender is associated with differing attitudes toward members of out-groups (e.g. Cummings, Kropf, & Weaver, 2000). Our ultimate goal is to contribute to ongoing efforts to identify a psychometrically sound scale that can be used to assess the effectiveness of interventions designed to improve attitudes toward PWD.

Methods

Data

Upon approval by the institutional review board and with the permission of individual course instructors, the Allophilia scale was administered at the beginning of the semester, during regularly scheduled class times, to students in undergraduate courses at a state university in the Midwest. Turning in a completed questionnaire constituted consent to participate. Data were collected in both introductory and advanced courses; specifically, undergraduate gerontology courses and other courses including sociology, educational psychology, kinesiology, family studies, and nursing.

In addition to completing the Allophilia scale, students provided basic demographic and descriptive information (i.e. age (in years), year in college (freshman, sophomore, junior, or senior/and above), gender, prior experience with PWD (see Table 1 for the specific questions)), which is known to be associated with differing attitudes toward members of out-groups (e.g. Cummings et al., 2000). Table 1 provides descriptive data on participants' demographic characteristics and experiential variables. Given possible effects on attitudes toward PWD (e.g. Yamashita et al., 2013), 34 students who had participated in a service learning program with PWD prior to the data collection period were excluded. Also, 19 students who completed multiple surveys (due to being enrolled in more than one course in which data collection occurred; the earliest one completed was retained) and 14 students who turned in incomplete questionnaires were excluded from analysis. The final sample size was 465 participants. The sample size is sufficient for confirmatory factor analysis (CFA) according to the existing guidelines of at least 10 cases per freely estimated parameter (Schreiber, Nora, Stage, Barlow, & King, 2006). This study adopted the full information maximum likelihood estimation technique which is appropriate to the data distributions in this study (DeMaris, 2004; Rhemtulla, Brosseau-Liard, & Savalei, 2012) and is capable of estimating parameters with missing values in the model (Arbuckle, 1996).

Table 1.

Participants' demographic characteristics.

	Total (n = 465)	Females (n = 342)	Males (n = 123)
	M (SD)	M (SD)	M (SD)
Age (years)	20.0 (2.10)	19.9 (2.26)	20.2 (1.65)
Current year in college *
Freshman	28.2%	30.4%	22.0%
Sophomore	21.9%	20.8%	25.2%
Junior	24.9%	23.7%	28.5%
Senior or higher	24.9%	25.2%	24.4%
Any personal experience interacting with family members or friends who have dementia? (Yes)	37.0%	36.5%	38.2%
Any volunteer or work experience at nursing homes or intergenerational programs? (Yes)	41.7%	45.9%***	30.1%
Any volunteer or work experience with PWD? (Yes)	17.6%	19.3%	13.0%
Courses **
Introductory gerontology	42.2%	42.7%	43.1%
Advanced gerontology	12.3%	15.5%	5.6%
Sociology	17.9%	16.1%	22.9%
Kinesiology and nursing	5.6%	5.6%	5.7%
Family study and educational psychology	20.9%	20.2%	22.8%

p < 0.05; **p < 0.01; ***p < 0.001(t test for age, and x² test for others—females versus males).

Measures—The Allophilia scale

The Allophilia scale was developed by Pittinsky et al. (2011b) to measure positive (rather than negative) attitudes toward members of out-groups. Whereas most attitudinal scales measure negative attitudes, Pittinsky et al. (2011a) maintain that individuals feel more comfortable reporting positive attitudes rather than negative attitudes and, in intervention work, identification of positive attitudes provides a foundation upon which prosocial relationships can be built between in- and out-group members. The Allophilia scale consists of 17 questions (see Table 2) that comprise five domains: (1) affection (i.e. “positive affective evaluations of out-group members”), (2) comfort (i.e. “a feeling of ease with out-group members”), (3) kinship (i.e. “a feeling of closeness with out-group members”), (4) engagement (i.e. “a tendency to seek to affiliate and interact with out-group members”), and (5) enthusiasm (i.e. “having emotionally heightened positive attitudes about out-group members”) (Pittinsky et al., 2011b, p. 46). Each item is rated on a 6-point Likert scale ranging from (1) strongly disagree to (6) strongly agree, with greater scores indicating more positive attitudes.

Table 2.

Model fit indices and standardized regression weights in CFA (Model 1 and Model 2) and multigroup CFA by gender.

Item	Dimension	Item	Standardized regression weights
						Multigroup analysis^a
			Model 1^b (all) N = 465	Model 2^b (all) N = 465	Model 3 (all) N = 465	Model 3^b (female) N = 342	Model 3^b (Male) N = 123
	Attitudes toward PWD	Affection	–	–	.819***	–	–
		Comfort	–	–	.841***	–	–
		Kinship	–	–	.831***	–	–
		Engagement	–	–	.757***	–	–
		Enthusiasm	–	–	.786***	–	–
1	Affection	In general, I have positive attitudes about PWD	.847***	.847***	.842***	.847***	.880***
2	Affection	I respect PWD	.676***	.676***	.672***	.676***	.677***
3	Affection	I like PWD	.827***	.827***	.823***	.846***	.779***
4	Affection	I feel positively toward PWD	.884***	.901***	.903***	.900***	.911***
5	Comfort	I am at ease around PWD	.907***	.907***	.907***	.922***	.855***
6	Comfort	I am comfortable when I hang out with PWD	.963***	.963***	.965***	.969***	.939***
7	Comfort	I feel like I can be myself around PWD	.683***	.683***	.679***	.697***	.632***
8	Kinship	I feel a sense of belonging with PWD	.897***	.900***	.893***	.923***	.819***
9	Kinship	I feel a kinship with PWD	.925***	.921***	.927***	.939***	.865***
10	Kinship	I would like to be more like PWD	.504***	–	–	–	–
11	Engagement	I am truly interested in understanding the points of view of PWD	.746***	.746***	.763***	.751***	.717***
12	Engagement	I am motivated to get to know PWD better	.881***	.881***	.889***	.878***	.881***
13	Engagement	To enrich my life, I would try and make more friends who are PWD	.794***	.793***	.778***	.803***	.734***
14	Engagement	I am interested in hearing about the experiences of PWD	.704***	.704***	.701***	.721***	.628***
15	Enthusiasm	I am impressed by PWD	.854***	.855***	.857***	.861***	.828***
16	Enthusiasm	I feel inspired by PWD	.868***	.869***	.878***	.854***	.905***
17	Enthusiasm	I am enthusiastic about PWD	.829***	.828***	.817***	.831***	.805***
		x² (DF)	413.05 (109)***	325.76 (94)***	449.60 (99)***
		CFI^b	.950	.957	.938	.954
		TLI^b	.937	.945	.925	.941
		RMSEA^b	.076	.074	.087	.055

CFI: comparative fit index; RMSEA: root mean square error of approximation; TLI: Tucker–Lewis index.

*< .05;

**< .01;

***

<.001.

The nested model evaluation showed that no statistically significant difference in the regression weights between female and male.

Model 1 includes all 17 items; Model 2 includes 16 items (item 10 was eliminated).

Pittinsky et al. (2011a) initially tested the Allophilia scale, using African American as out-group members, with a large sample of undergraduate students, and reported good psychometric properties (root mean square error of approximation (RMSEA) = .06; comparative fit index (CFI) = .95 in a CFA; Cronbach's alphas ranged from .88 to .92). The Allophilia scale has subsequently been validated with racial minorities and can be flexibly applied for a variety of out-groups (Pittinsky et al., 2008). Indeed, to date, the Allophilia scale has been used to assess attitudes toward other out-groups members such as LGBT individuals (Fingerhut, 2011) and foreigners (Alfieri & Marta, 2011). In the current study, the specified out-group was “PWD.” To our knowledge, this research is the first to use the Allophilia scale to measure attitudes toward this particular out-group.

Analytic strategy

We employed CFA to examine the psychometric properties of the Allophilia scale using AMOS 17.0 (Arbuckle, 2008). The main objective of CFA is to evaluate whether the observed data fit a hypothesized/theoretical model (in this case the five domains in the Allophilia scale: affection, comfort, kinship, engagement, and enthusiasm) (Brown, 2006). We conducted the CFA in five steps: (1) model specification, (2) model identification, (3) model estimation, (4) model evaluation, and (5) model respecification.

First, the model was specified based on the work of Pittinsky et al. (2011b). Two domains, affection and engagement, each contained four items; the remaining domains each contained three items (Model 1—see Figure 1). In Pittinsky et al. (2011a), the five Allophilia scale dimensions were highly correlated (Pearson's r ranged from .62 to .78) and they are reflected in the first-order CFA model in this study. We also examined a second-order CFA model, which includes the higher order latent variable (i.e. attitudes toward PWD) predicting the five domains. Such higher order latent variables represent correlations between the first-order latent variables (i.e. the five domains). Second, we examined the model identification that is a required condition for parameter estimation in CFA. Simply put, a model is identified when the number of known pieces of information (i.e. variances and covariances of observed variables) in the data is equal (just identified) or greater than (over identified) the number of freely estimated parameters in a model (Bollen, 1989). Based on the empirical testing function built into AMOS, we verified that all models were over-identified. Third, the full information maximum likelihood estimation method was used to estimate parameters in the model. As a fourth step, we evaluated the model based on three criteria: the Tucker–Lewis index (TLI), CFI, and RMSEA. A TLI and CFI close to or greater than .95 and a RMSEA close to .06 or below indicate good fit; values greater than .90 and .10 or below, respectively, indicate acceptable fit (Brown, 2006). In addition, given reported gender differences in attitudes toward PWD in a meta-analysis (Kite et al., 2005), we ran multigroup CFA and invariance tests by gender. The invariance tests compare each parameter estimate (e.g. regression weight) between females and males. Finally, in the fifth step of model respecification, we reviewed the estimated standardized regression coefficients and reliability (Cronbach's alpha) for all scale items and correlations between factors. In addition, each scale item was also reexamined to determine whether it was suitable in the context of attitudes toward PWD.

Figure 1.

(a, b) Model 1: first-order confirmatory factor analysis (CFA) and Model 2: modified first-order CFA. (c) Model 3: second-order CFA. Notes: (a) Model 2 excludes item 10. (b) The shown regression weights are for Model 2 (see Table 2 for the parameter estimates of Model 1). (c) Model 3 is based on Model 2 (excluding item 10).

Results

The characteristics of the respondents are summarized in Table 1; data are presented for the whole sample and for females (n = 342) and males (n = 123) separately. The mean age of the respondents was 20.2 years old (SD = 2.1). Although there were few differences between female and male participants with respect to many of the demographic and experiential characteristics, females were more likely to have volunteer or work experience at nursing homes and/or participated in intergenerational programs than males. A significantly higher proportion of female than male participants were first year students (30% versus 22%; x² (17, N = 465) = 11.72, p < 0.05). As can be seen in Table 1, approximately 42% of the total participants were enrolled in introductory gerontology courses. A significantly higher proportion of female than male students also were enrolled in advanced gerontology courses (15.5% versus 5.6%, respectively; x² (5, N = 465) = 35.45, p < 0.01).

Table 2 shows the model fit indices, the estimated standardized regression coefficients in Model 1, Model 2, and multigroup CFA for the females and males. Model 1 replicated the model specification in Pittinsky et al. (2011b). Model 1 showed a moderate fit (x² (109, N = 465) = 413.02; CFI = .954; TLI = .935; RMSEA = .070). The chi-square statistic was not taken into account because it returns a statistically significantly result (i.e. poor fit) for a negligible difference in a large sample size (Brown, 2006). Therefore, we only examined the consistency of CFI, TLI, and RMSEA—the three recommended model fit indices. The majority of items (16 items) showed high standardized regression weights that ranged from .704 to .954. However, one item, “I would like to be more like PWD,” showed a considerably lower weight (λ = .492) compared to the others. Conceptually, this item is not suitable for the Allophilia scale in the context of dementia because most individuals do not aspire to develop dementia, presumably due to ageism-based fear (Kitwood, 1990). Therefore, we excluded this item from the Allophilia scale. After excluding this item (Model 2), the model fit improved (x² (94, N = 465) = 325.76; CFI = .964; TLI = .947; RMSEA = .065). In addition, all standardized regression weights were consistently high. As such, based on empirical grounds, we decided to adopt Model 2. The reliability scores (Cronbach's alphas) ranged from .882 and .914; all alphas are reported in Table 3.

Table 3.

The correlations between five factors in the Allophilia scale by gender.^a

	Affection		Comfort		Kinship		Engagement
	Female	Male	Female	Male	Female	Male	Female	Male
Affection (.897)^b	–	–
Comfort (.894)^b	.775	.638	–	–
Kinship (.903)^b	.692	.539	.821	.636	–	–
Engagement (.900)^b	.660	.449	.615	.426	.655	.631	–	–
Enthusiasm (.894)^b	.731	.575	.647	.482	.683	.567	.801	.756

All correlations (Pearson's r) were statistically significant (p < .05).

Cronbach's alpha.

Note: The correlations are based on Model 2 (see “Methods” section).

With regards to the invariance test, the female–male model comparisons showed no statistically significant differences between the unconstrained model and constrained model (i.e. the regression weights are forced to be equal between females and males) of measurement weights (factor loadings) (x² (11, N = 465) = 13.10, p = .283). Therefore, the modified Allophilia scale (Model 2) is valid both for males and females. However, examination of the correlations between factors revealed several gender differences (see Table 3). Specifically, our follow-up analyses suggested that there are no significant gender differences in the correlations between enthusiasm and engagement, enthusiasm and kinship, and engagement and kinship, whereas there were significant differences in the other correlations between factors as a function of gender. For instance, the correlations between affection and engagement (r _female = .660 versus r _male = .449) are considerably different compared to other observed correlations (see Table 3). That is, when positive attitudes in the affection domain are observed, females are more likely to have positive attitudes in the engagement domain than males.

Building on Model 2, the higher order CFA (Model 3) was evaluated to assess the detailed associations between the indicators (i.e. Allophilia scale items), first-order latent variables (i.e. five domains), and second-order latent variable (i.e. attitude toward PWD) (see Figure 1). Overall, the model showed acceptable fit (x² (99, N = 465) = 449.60; CFI = .938; TLI = .925; RMSEA = .087). All five domains loaded strongly onto “attitudes toward PWD”; the standardized regression weights ranged from .757 to .841. Thus, we conclude that both the first-order (Model 2) and second-order (Model 3) CFA showed reasonable fit and could be interchangeably used according to the purpose of a specific assessment of attitudes toward PWD (e.g. assessing overall attitudes as opposed to specific attitudinal domains).

Discussion

The purpose of this research was to identify a psychometrically and conceptually sound scale to assess positive attitudes toward people with dementia. Our results indicate that, after removing one item, the Allophilia scale (Pittinsky et al., 2011a, 2011b) is such a scale for use with undergraduate students. This finding contributes to a growing body of research that validates the use of the Allophilia scale to assess positive attitudes (i.e. affection, comfort, kinship, engagement, enthusiasm) toward members of diverse out-groups, including African Americans, Latinas/Latinos, and members of the LGBT community in the U.S. (e.g. Fingerhut, 2011; Pittinsky et al., 2011b) and Romanians in Italy (Alfieri & Marta, 2011).

Young adults' attitudes toward older adults have been a focus of gerontological research for more than five decades, and the use of the Allophilia scale in aging-related attitudinal research is innovative in several ways. First, the long-standing tradition of measuring negative as opposed to positive attitudes toward older adults in general and those with dementia appears to be changing. In addition to our application of the Allophilia scale to people with dementia, O'Connor and McFadden (2010) developed the Dementia Attitudes scale (DAS), which comprises two factors: individuals' knowledge about dementia and social comfort with people who have the disease. Although both the DAS and Allophilia scales require additional use with different samples to confirm their validity and reliability, these scales have the potential to document positive attitudes toward individuals with dementia. Second, both the Allophilia and Dementia Attitude scales were developed with attention to psychometric rigor, whereas the psychometric properties of many of the most widely used tools that measure (negative) attitudes toward older adults (e.g. Kogan Attitudes toward OPS, ASD scale, Facts on Aging quizzes) have fairly serious psychometric limitations (Cowan et al., 2004; Iwasaki & Jones, 2008).

There are both conceptual and practical reasons to assess positive attitudes toward members of an out-group. First, because positive and negative attitudes are neither theoretically nor empirically the exact opposite of each other (Pittinsky, 2009; Pittinsky et al., 2011a), the multidimensionality of complex attitudes will not be captured without assessing attitudes of both valences. Second, given that reducing negative attitudes is necessary, but not sufficient, to promote positive out-group relations (Pittinsky et al., 2008, 2011a, 2011b), and that positive attitudes foster proactive and prosocial behaviors (Pittinsky, 2009), documentation of positive attitudes can be particularly important when attempting to evaluate interventions to promote positive attitudes toward members of an out-group (Pittinsky et al., 2011a).

Two contemporary theories of ageism, social identity theory (Tajfel, 1981) and terror management theory (Greenberg et al., 1997; Solomon et al., 2004), conceptualize younger adults' ageism toward older adults in terms of in-group and out-group relations. Viewed within this framework, ageism serves to marginalize members of the out-group. This marginalization has been described by a variety of terms, including othering (e.g. Bytheway, 2005; Jönson, 2013), malignant social psychology (Kitwood, 1990), and malignant social positioning (Sabat, 2003). Regardless of the term that is used, the consequences of ageism are sufficiently severe that it has been characterized as a social issue (Behuniak, 2011) and civil rights/social justice issue that spreads across individuals and generations and should not be tolerated (Palmore, 2015). As such, Palmore calls for “evidence-based interventions that sucessfully and efficiently reduce ageism” (p. 873) and “proven methods of reversing ageism” (p. 873). This is especially important in light of the projected increase in the number of older adults with dementia and their corresponding need for informal and formal care.

To date, with few exceptions, documentation of students' attitudes and evaluation of educational interventions to reduce/reverse ageism, whether classroom-based and/or experiential, overwhelmingly focus on lessening students' negative attitudes (e.g. Kaf et al., 2011; Stuart-Hamilton & Mahoney, 2003; Yamashita et al., 2013). The inclusion of positive attitudes, as assessed by the Allophilia scale, might provide useful information about previously ignored increases in participants' positive attitudes and aid in the interpretation of mixed attitudinal findings (Allan & Johnson, 2008; Palmore, 1998; Yamashita et al., 2013). Although documentation of attitudes in and of itself is of conceptual interest, increasingly researchers are focusing on interventions to promote more prosocial behavior among young adults with respect to older adults with dementia. The explicit identification of positive attitudes can serve as a foundation for these efforts.

In her discussion of how to resist extreme ageism (or the “zombie-fication”) of individuals with dementia, Behuniak (2011) emphasized the importance of connectedness, commonality, and interdependency between individuals with and without the disease. These qualities, along with a commitment to provide a failure-free experience, are critical components of an emerging group of creative arts-based interventions for individuals with dementia. Many of these interventions are designed to be/involve intergenerational partnerships between individuals with dementia and students/young adults.

A growing body of literature demonstrates the benefits older adults with dementia receive from participating in creative arts-based programs such as Memories-in-the-Making® (http://www.alz.org/oc/in_my_community_10849.asp; e.g. Kinney & Rentz, 2005; Rentz, 2002); Opening Minds through Art (http://miamioh.edu/cas/academics/centers/scripps/community-connections/opening-minds-through-art/index.html; e.g. Sauer, Fopma-Loy, Kinney, & Lokon, 2014); and TimeSlips, a group storytelling program (http://www.timeslips.org/; e.g. George & Houser, 2014; Phillips, Reid-Arndt, & Pak, 2010). Additional research (e.g. George, Stuckey, & Whitehead, 2014; Lokon et al., 2012; Yamashita et al., 2013) documents the benefits that young adult participants derive from these intergenerational programs. The Allophilia scale has the potential to help identify interventions that are especially successful at increasing young adults' positive attitudes toward individuals with dementia, which could serve as a springboard to increase meaningful interaction, productive interdependency, and decrease young adults' ageism toward individuals with dementia.

This study is not without limitations. First, although two-fifths (42%) of the participants were students in an introductory gerontology course, it should not be assumed that these students were particularly interested in gerontology. Because this course fulfills a liberal education and social science requirement, many students who enroll in the course are not particularly interested in the topic, and some do not even know that gerontology is the study of aging. Having said that, 12% of the samples were students enrolled in an advanced gerontology course. It is possible that these students were both more knowledgeable about, and had more favorable attitudes toward, individuals with dementia. In future research as well as practice, pre- and post-intervention (e.g. gerontology course, service learning) use of the Allophilia scale is recommended to identify differing effects on positive attitudes toward PWD across subgroups (e.g. introductory versus advanced gerontology students). Second, because the course in which each student was enrolled prior to this study potentially constitutes an intervention, we do not have any information on the type and dose of such intervention. Presumably, students were more likely to be exposed to information on dementia in a gerontology than in a nongerontology course. Even within different sections of the same introductory gerontology course, there is variability in the amount and type of information that is presented on dementia prior to data collection.

Third, although there is some evidence that the Allophilia scale can be used to measure the attitudes of middle aged and old adults toward different out-groups (Alfieri & Marta, 2011; Fingerhut, 2011), our findings should not be extended to other nonuniversity student populations without further validation. Although our study verified that the Allophilia scale can be used to measure female and male undergraduate students' attitudes, the scale has yet to be tested for other subgroups. On a related note, it is critical to pay close attention to correlations between attitudinal domains as demonstrated in Table 3 because they may reveal reasons why subgroup differences exists (e.g. interactions between domains) in future research. Finally, the reliability and validity of the Allophilia scale may benefit from further research with data collected over a longer period of time.

In summary, our preliminary validation of the Allophilia scale in the context of attitudes toward people with dementia enables the assessment of positive domains of students' attitudes (i.e. affection, comfort, kinship, engagement, enthusiasm) that typically are overlooked, despite their potential importance in promoting proactive and prosocial behavior toward out-group members (Pittinsky & Montoya, 2009; Pittinsky et al., 2011a), the goal of any intervention. This bodes well for future interventions that are designed to prepare informal networks of family members and friends and formal care providers to meet the emotional, social, and health needs of the growing numbers of individuals with dementia.

Footnotes

Acknowledgements

The authors thank Dr Elizabeth Lokon for helpful comments on an earlier version of this article.

Author Note

Jennifer M Kinney and Takashi Yamashita are the co-first authors.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Jennifer M Kinney, PhD, is a professor of gerontology in the Department of Sociology and Gerontology and a research fellow with the Scripps Gerontology Center at Miami University, Oxford, Ohio. Her primary research focuses on families who have a relative with dementia. Recent work includes the evaluation of creative arts and Montessori-based programs designed to promote quality of life among individuals with dementia and the use of technology to facilitate inhome family caregiving. A secondary interest is the impact of intergenerational/service learning programs on college students’ attitudes toward older adults, including those with dementia.

Takashi Yamashita, PhD is an assistant professor of sociology at University of Nevada Las Vegas (UNLV). His research interests are geographic access to health resources, including health care services, healthy food outlets and places for physical activities, health literacy over the life course, fall risk factors and chronic disease self-management behaviors in older populations. He has been involved in other research areas such as measurement for an international comparisons of health literacy, quality of life in Eastern Asian populations, psychometric property of residential care facility residents’ satisfaction, effects of marital quality on health, scientific data visualization, and exploratory spatial analysis of public health policy and gerontology education.

J Scott Brown, PhD, is an associate professor of sociology and gerontology and a research fellow with Scripps Gerontology Center at Miami University, Oxford, Ohio. His research interests are focused on wealth and health inequalities across the life course with particular emphasis on gender and race differences in physical and mental health. Dr Brown has published numerous research articles and book chapters in these areas as well as coauthoring a book on global aging. He also serves as an associate editor for the Journals of Gerontology: Social Sciences.

References

Alfieri

Marta

(2011) Positive attitudes towards the outgroup: Adaptation and validation of the allophilia scale. Testing, Psychometrics, Methodology in Applied Psychology 18(2): 99–116.

Allan

L. J.

Johnson

J. A.

(2008) Undergraduate attitudes toward the elderly: The role of knowledge, contact and aging anxiety. Educational Gerontology 35(1): 1–14. DOI:10.1080/03601270802299780.

Arbuckle

J. L.

(2008) Amos 17.0.0, Crawfordville, FL: Amos Development.

Arbuckle, J. L. (1996). Full information estimation in the presence of incomplete data. In G. A. Marcoulides & R. E. Schumacker (Eds.), Advanced structural equation modeling: Issues and techniques. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Behuniak

S. M.

(2011) The living dead? The construction of people with Alzheimer's disease as zombies. Ageing and Society 31(01): 70–92.

Blieszner, R., & Artale, L. M. (2001). Benefits of intergenerational service-learning to human service majors. Educational Gerontology, 27, 71–87.

Bodner

(2009) On the origins of ageism among older and younger adults. International Psychogeriatrics 21(06): 1003–1014. DOI:10.1017/S104161020999055X.

Bollen

K. A.

(1989) Structural equations with latent variables, New York, NY: Wiley.

Breytspraak

L. M.

Arnold

Hogan

(2008) Dimensions of an intergenerational relationship between medical students and mentors-on-aging. Journal of Intergenerational Relationships 6(2): 131–153. DOI:10.1080/15350770801955040.

10.

Brookmeyer

Evans

D. A.

Hebert

Langa

K. M.

Heeringa

S. G.

Plassman

B. L.

Kukull

W. A.

(2011) National estimates of the prevalence of Alzheimer's disease in the United States. Alzheimer's and Dementia 7(1): 61–73. DOI:https://dx-doi-org.web.bisu.edu.cn/10.1016/j.jalz.2010.11.007.

11.

Brown

T. A.

(2006) Confirmatory factor analysis for applied research, New York: The Guilford Press.

12.

Butler

R. N.

(1975) Why survive? Being old in America, Oxford, England: Harper & Row.

13.

Butler

R. N.

(1980) Ageism: A foreword. Journal of Social Issues 36(2): 8–11.

14.

Bytheway

(2005) Ageims. In: Johnson

M. L.

Coleman

Bengtson

(eds) The Cambridge handbook on age and ageing, Cambridge, UK: Cambridge University Press, pp. 338–345.

15.

Chung

J. C. C.

(2009) An intergenerational reminiscence programme for older adults with early dementia and youth volunteers: Values and challenges. Scandinavian Journal of Caring Sciences 23(2): 259–264. DOI:10.1111/j.1471-6712.2008.00615.x.

16.

Cowan

D. T.

Fitzpatrick

J. M.

Roberts

J. D.

While

A. E.

(2004) Measuring the knowledge and attitudes of health care staff toward older people: Sensitivity of measurement instruments. Educational Gerontology 30(3): 237–254. DOI:10.1080/03601270490273169.

17.

Cummings

S. M.

Kropf

N. P.

Weaver

K. L. D.

(2000) Knowledge of and attitudes toward aging among non-elders: Gender and race differences. Journal of Women and Aging 12(1–2): 77–91. DOI:10.1300/J074v12n01_06.

18.

DeMaris

(2004) Regression with social data: Modeling continuous and limited response variables, Hoboken, NJ: John Wiley & Sons, Inc.

19.

Fingerhut

A. W.

(2011) Straight allies: What predicts heterosexuals' alliance with the LGBT community? Journal of Applied Social Psychology 41(9): 2230–2248. DOI:10.1111/j.1559-1816.2011.00807.x.

20.

Fraboni

Saltstone

Hughes

(1990) The Fraboni Scale of Ageism (FSA): An attempt at a more precise measure of ageism. Canadian Journal on Aging/La Revue canadienne du vieillissement 9(01): 56–66. DOI:10.1017/S0714980800016093.

21.

George

D. R.

Houser

W. S.

(2014) “I'm a Storyteller!”: Exploring the benefits of timeslips creative expression program at a nursing home. American Journal of Alzheimer's Disease and Other Dementias 29(8): 678–684. DOI:10.1177/1533317514539725.

22.

George

D. R.

Stuckey

H. L.

Whitehead

M. M.

(2014) How a creative storytelling intervention can improve medical student attitude towards persons with dementia: A mixed methods study. Dementia 13(3): 318–329. DOI:10.1177/1471301212468732.

23.

Greenberg

Solomon

Pyszczynski

(1997) Terror management theory of self-esteem and cultural worldviews: Empirical assessments and conceptual refinements. In: Zanna

M. P.

(ed.) Advances in experimental social psychology Vol 29, San Diego, CA: Academic Press, pp. 61–139.

24.

Iwasaki

Jones

J. A.

(2008) Attitudes toward older adults: A reexamination of two major scales. Gerontology and Geriatrics Education 29(2): 139–157. DOI:10.1080/02701960802223209.

25.

Jones

E. E.

(1998) Major developments in five decades of social psychology. In: Gilbert

D. T.

Fiske

S. T.

Lindzey

(eds) The handbook of social psychology Vol 1, 4th ed. Boston, MA: McGraw-Hill, pp. 3–57.

26.

Jönson

(2013) We will be different! Ageism and the temporal construction of old age. The Gerontologist 53(2): 198–204. DOI:10.1093/geront/gns066.

27.

Kaf

W. A.

Barboa

L. S.

Fisher

B. J.

Snavely

L. A.

(2011) Effect of interdisciplinary service learning experience for audiology and speech-language pathology students working with adults with dementia. American Journal of Audiology 20(2): S241–S249.

28.

Kinney

J. M.

Rentz

C. A.

(2005) Observed well-being among individuals with dementia: Memories in the making (C), an art program, versus other structured activity. American Journal of Alzheimer's Disease and Other Dementias 20(4): 220.

29.

Kite

M. E.

Stockdale

G. D.

Whitley

B. E.

Johnson

B. T.

(2005) Attitudes toward younger and older adults: An updated meta-analytic review. Journal of Social Issues 61(2): 241–266. DOI:10.1111/j.1540-4560.2005.00404.x.

30.

Kitwood

(1990) The dialectics of dementia: With particular reference to Alzheimer's disease. Ageing and Society 10(02): 177–196. DOI:10.1017/S0144686X00008060.

31.

Kogan

(1961) Attitudes toward old people: The development of a scale and an examination of correlates. The Journal of Abnormal and Social Psychology 62(1): 44–54. DOI:10.1037/h0048053.

32.

Lokon

Kinney

J. M.

Kunkel

(2012) Building bridges across age and cognitive barriers through art: College students' reflections on an intergenerational program with elders who have dementia. Journal of Intergenerational Relationships 10(4): 337–354. DOI:10.1080/15350770.2012.724318.

33.

Martens

Goldenberg

J. L.

Greenberg

(2005) A terror management perspective on ageism. Journal of Social Issues 61(2): 223–239. DOI:10.1111/j.1540-4560.2005.00403.x.

34.

Martens

Greenberg

Schimel

Landau

M. J.

(2004) Ageism and death: Effects of mortality salience and perceived similarity to elders on reactions to elderly people. Personality and Social Psychology Bulletin 30(12): 1524–1536. DOI:10.1177/0146167204271185.

35.

Nelson

T. D.

(2004) Ageism: Stereotyping and prejudice against older persons, Boston, MA: MIT press.

36.

Nelson

T. D.

(2009) Ageism. In: Nelson

T. D.

(ed.) Handbook of prejudice, stereotyping, and discrimination, New York, NY: Taylor and Francis, pp. 431–440.

37.

O'Connor

M. L.

McFadden

S. H.

(2010) Development and psychometric validation of the dementia attitudes scale. International Journal of Alzheimer's Disease. DOI:10.4061/2010/454218.

38.

Palmore

E. B.

(1998) The facts on aging quiz, 2nd ed. New York, NY: Springer.

39.

Palmore

E. B.

(1999) Ageism: Negative and positive, New York: Springer Publishing Company.

40.

Palmore

E. B.

(2015) Ageism comes of age. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 70(6): 873–875. DOI:10.1093/geronb/gbv079.

41.

Phillips

L. J.

Reid-Arndt

S. A.

Pak

(2010) Effects of a creative expression intervention on emotions, communication, and quality of life in persons with dementia. Nursing Research 59(6): 417–425. DOI:10.1097/NNR.0b013e3181faff52.

42.

Pillemer

Schultz

(2002) Evaluation of the student assisted independent living (SAIL) service-learning project. In: Seperson

S. B.

Hageman

(eds) Elder care and service learning: A handbook, Westport, CT: Auburn House, pp. 252–260.

43.

Pittinsky

T. L.

(2009) Look both ways. Phi Delta Kappan 90(5): 363–364.

44.

Pittinsky

T. L.

Montoya

R. M.

(2009) Is valuing equality enough? Equality values, Allophilia, and social policy support for multiracial individuals. Journal of Social Issues 65(1): 151–163. DOI:10.1111/j.1540-4560.2008.01592.x.

45.

Pittinsky

T. L.

Ratcliff

J. F.

Maruskin

L. A.

(2008) Coexistence in Israel: A national study, Cambridge, MA: Harvard Kennedy School, Center for Public Leadership.

46.

Pittinsky

T. L.

Rosenthal

S. A.

Montoya

R. M.

(2011a) Liking is not the opposite of disliking: The functional separability of positive and negative attitudes toward minority groups. Cultural Diversity and Ethnic Minority Psychology 17(2): 134–143. DOI:10.1037/a0023806.

47.

Pittinsky

T. L.

Rosenthal

S. A.

Montoya

R. M.

(2011b) Measuring positive attitudes toward outgroups: Development and evaluation of the Allophilia scale. In: Tropp

L. R.

Mallett

R. K.

(eds) Moving beyond prejudice and reduction: Pathways to positive intergroup relationship, Washington, DC: American Psychological Association, pp. 41–60.

48.

Plassman

B. L.

Langa

K. M.

Fisher

G. G.

Heeringa

S. G.

Weir

D. R.

Ofstedal

M. B.

Rodgers

W. L.

(2007) Prevalence of dementia in the United States: The aging, demographics, and memory study. Neuroepidemiology 29(1–2): 125–132.

49.

Polizzi

K. G.

Millikin

R. J.

(2002) Attitudes toward the elderly: Identifying problematic usage of ageist and overextended terminology in research instructions. Educational Gerontology 28(5): 367–377. DOI:10.1080/03601270290081344.

50.

Rentz

C. A.

(2002) Memories in the Making©: Outcome-based evaluation of an art program for individuals with dementing illnesses. American Journal of Alzheimer's Disease and Other Dementias 17(3): 175–181.

51.

Rhemtulla

Brosseau-Liard

P. É.

Savalei

(2012) When can categorical variables be treated as continuous? A comparison of robust continuous and categorical SEM estimation methods under suboptimal conditions. Psychological Methods 17(3): 354–373. DOI:10.1037/a0029315.

52.

Rosencranz

H. A.

McNevin

T. E.

(1969) A factor analysis of attitudes toward the aged. The Gerontologist 9(1): 55–59. DOI:10.1093/geront/9.1.55.

53.

Sabat

S. R.

(2003) Malignant positioning and the predicament of people with Alzheimer's disease. In: Haeer

Moghaddam

F. M.

(eds) The self and others: Positioning individuals and groups in personal, political, and cultural contexts, Westport, CT: Praeger, pp. 85–98.

54.

Sauer, P. E., Fopma-Loy, J., Kinney, J. M., & Lokon, E. (2014). “It makes me feel like myself”: Person-centered versus traditional visual arts activities for people with dementia. Dementia DOI:10.1177/1471301214543958.

55.

Schreiber

J. B.

Nora

Stage

F. K.

Barlow

E. A.

King

(2006) Reporting structural equation modeling and confirmatory factor analysis results: A review. The Journal of Educational Research 99(6): 323–338. DOI:10.3200/JOER.99.6.323-338.

56.

Solomon

Greenberg

Pyszczynski

(2004) The cultural animal: Twenty years of terror management theory and research. In: Greenberg

Koole

S. L.

Pyszczynski

(eds) Handbook of experimental existential psychology, New York, NY: The Guilford Press, pp. 13–34.

57.

Stuart-Hamilton

Mahoney

(2003) The effect of aging awareness training on knowledge of, and attitudes towards, older adults. Educational Gerontology 29(3): 251–260. DOI:10.1080/713844305.

58.

Tajfel

(1981) Human groups and social categories: Studies in social psychology, New York, NY: Cambridge University Press.

59.

Yamashita

Kinney

J. M.

Lokon

E. J.

(2013) The impact of a gerontology course and a service-learning program on college students' attitudes toward people with dementia. Journal of Applied Gerontology 32(2): 139–163. DOI:10.1177/0733464811405198.