Validation of a Tool to Assess and Track Undergraduate Attitudes Toward Those Living in Poverty

Abstract

Purpose:

This article describes the development and validation of the Undergraduate Perceptions of Poverty Tracking Survey (UPPTS).

Method:

Data were collected from 301 undergraduates at a small university in the Northeast and analyzed using exploratory factor analysis augmented by random qualitative validation.

Results:

The resulting survey contains 39 questions and has six factors that meet empirical standards for validity and reliability. The UPPTS provides information regarding undergraduate students’ perceptions of those living in poverty in three areas: (1) general attitudes toward those living in poverty, including a sense of the students’ underlying explanation for why someone may be poor; (2) understanding of and empathy for those living in poverty; and (3) commitment to addressing poverty via direct action or support for programs/services that aid those in poverty.

Discussion:

The UPPTS builds upon the concepts of a lack of social empathy and cognitive distancing as principal reasons why people fail to do more to help the poor via either direct action or support for programs that will aid the poor. Further, social work and other educators may use the UPPTS to guide their efforts in poverty education and to track the progress of their efforts with undergraduate students.

Keywords

poverty survey tracking undergraduate education validation

Introduction and Review of Literature

Austin (2007, p. 1) has noted that addressing poverty seems to become an area of concern for the general public and for universities every few decades. He notes that events such as the Great Depression or the publication of an important book such as Smith’s (1943) A Tree Grows in Brooklyn or Harrington’s (1962) The Other America: Poverty in the United States bring an awareness and subsequently a desire to take action to the forefront of the American public’s agenda. In the aftermath of hurricane Katrina and the publicity that documented the poverty faced by so many of the hurricane’s victims, addressing and teaching about poverty seems to have once again become a priority for Americans and for American universities. Such a priority is appropriate considering evidence provided by Buss (2010) that the average poor person has become more impoverished over the last two decades. However, as Austin (2007, p. 1) has also noted that the lack of an integrated approach to teaching students about poverty at American universities typically leads to a situation where “everybody’s business becomes nobody’s business.” That is, faculty and administration acknowledge the importance of educating about poverty, but in reality offer only a diffuse, scattered, and unfocused approach (Blair, 2010).

In addition to the curriculum challenges, universities and their faculty who want to assess the impact of their programs on the knowledge base, perceptions, and attitudes of undergraduates toward those living in poverty have not had a reliable, empirically validated instrument that would enable them to track the impact of their efforts. The Undergraduate Perceptions of Poverty Tracking Survey (UPPTS), an updated and expanded version of the Attitudes about Poverty and Poor People originally developed by Atherton and his colleagues (1993), is designed to address the need for such a tracking tool and to provide educators with the information they need about their students and about the impact of their educational efforts in the area of poverty (Blair, 2010).

Empathy, Cognitive Distancing, and Poverty

The importance of empathy in relation to the willingness to aid the poor is a theoretical construct that is gaining acceptance from poverty educators (Lane, 2001; Lott, 2002; Segal, 2007). Lane (2001) in particular has been both forceful and elegant in his discussion of the relationship between a lack of empathy for those living in poverty, an emphasis on self-reliance, and the tendency in the United States to separate one’s personal fate from the fate of the nation as a whole to culminate in the United States being one of the least egalitarian societies among developed nations. Lane argues that most Americans view the United States as an essentially fair place where individuals control their own destiny via the choices they make. Thus, poverty is attributed to personal failures rather than to a failure of society. Further, he believes that the nonpoor view the poor as receiving their “just deserts” and feel angry toward the poor for not being more successful. In contrast, those who live in non-Western societies seem more inclined to support structural explanations of poverty (Abouchedid & Nasser, 2001; Nasser, 2007).

As argued by Segal (2007, p. 72), a focus on empathy suggests “the differences and gaps between those who are well-off and those who are poor perpetuates poverty and keeps society from addressing the problems.” Segal also notes that this approach is the least likely to be taught and used as a way of getting students to engage in the process of alleviating poverty. In a similar vein, Lott (2002) argues that the principal response in the United States of the nonpoor to the poor is that of,

…distancing, that is, separation, exclusion, devaluation, discounting, and designation as “other,” and that this response can be identified in both institutional and interpersonal contexts. In social psychological terms, distancing and denigrating responses operationally define discrimination. These, together with stereotypes (i.e., a set of beliefs about a group that are learned early, widely shared, and socially validated) and prejudice (i.e., negative attitudes) constitute classism. (p. 100)

Clark (2007) also points to the body of research, especially in relation to social work students, that demonstrates that positive attitudes toward the poor are directly linked to exposure and direct experience with people who live in poverty; that is, direct exposure and experience with the poor breaks down the distancing and separation that Lott describes. The opportunity to build relationships with people living in poverty not only enhances understanding of poverty but also inhibits apathy (Zygmunt-Fillwalk, 2009). Having a positive view of those in poverty is linked to a willingness to work with the poor, to supporting efforts to improve the lives of the poor, and to efforts to eradicate poverty (Coryn, 2002; Cozzarelli, Wilkinson, & Tagler, 2001). Segal (2007) refers to this as the development of social empathy and questions the use of teaching and related materials that create sympathy rather than empathy for those in poverty.

Theoretical Considerations and the Need for an Improved Tracking Tool

The authors began with a series of discussions (many of these involving other colleagues from the university) that raised questions regarding the impact of class activities, such as service learning, reflection papers, and community projects on the students enrolled in classes that had strong poverty content. The primary question being: Do these activities raise the level of empathy and commitment of the students toward those living in poverty? Further, the authors have decided to add participation in a poverty simulation class activity, and while seeking information regarding the impact of the simulation, the authors came to realize that empirical assessments of simulations and other class activities in relation to poverty were minimal in the literature. It was a desire to empirically assess the impact of this simulation, as well as other educational efforts (e.g., service learning) that led the authors to develop the UPPTS.

The authors began a search for an empirically based instrument that could be used to assess the impact of their educational efforts (Coryn, 2002; Cozzarelli et al., 2001). The Attitudes Toward Poverty and Poor People scale introduced by Atherton et al. (1993) and originally directed toward social work educators was the closest fit the authors could locate. As described by Atherton et al., this instrument was developed because:

Social work researchers need a dependable measure of attitudes toward poverty and the poor population. In addition to research applications, social workers could use a dependable scale to identify pertinent issues in professional workshops that deal with attitudes toward poverty and poor people. Social work educators could use the scale to assess student attitudes toward the poor population when they are addressing value questions associated with poverty. (p. 1)

The Atherton scale has been utilized by researchers in a variety of ways. Rehner, Ishee, Salloum, and Velasques (1997) used the Atherton scale to measure 186 Mississippi social workers’ attitudes toward poverty. To gauge nursing students’ beliefs about the relationship between poverty and health, Reutter, Sword, Meagher-Stewart, and Rideout (2004) developed a 59-item questionnaire that incorporated the 37-item Atherton scale. In addition to demographic questions and items about exposure to poverty that were written for the study, the questionnaire included items from a survey about poverty and health designed by Reutter, Neufeld, and Harrison (1999). Proctor and colleagues (2010) also used the Atherton scale in conjunction with the survey tool from Reutter et al. (1999). They used the measures as a pretest/posttest to assess the effect of an experiential learning activity on students’ attitudes and beliefs about poverty and health. Gumpert and Kraybill-Greggo (2005) administered the Atherton questionnaire as a pretest/posttest to measure change in attitudes among undergraduates who participated in intensive service trips that involved working directly with impoverished people. To evaluate the impact of social work courses on attitudes toward poverty, Weaver and Yun (2011) administered the Atherton scale to 166 undergraduate students enrolled in three different courses at a Canadian university. They modified 2 items in the scale, replacing the phrase “food stamps” with “welfare benefits” to use language befitting the Canadian welfare state.

Although a good starting point, the Atherton scale was lacking in several areas. Most specifically, the scale does not assess the respondent’s empathy toward those living in poverty nor does it attempt to assess the respondent’s commitment to doing something that will aid the poor. Other researchers, such as Yun and Weaver (2010) have noted the deficiencies of the Atherton scale and have sought to make improvements. Yun and Weaver have developed a 21-item short form of the Atherton scale and suggest it is easier to administer and thus more efficient for data collection than the original 37-item scale. While Yun and Weaver provide a useful alternative to the Atherton scale (e.g., Patterson and Hulton (2012) used the short form to evaluate attitudes of students who participated in a poverty simulation), deficiencies persist.

Shek (Shek, 2004; Shek & Ma, 2009) in particular has noted the problems with existing instruments that attempt to measure respondents’ attitudes toward the poor and their underlying explanation for the causes of poverty. Shek (2004) states,

“Although many measures have been used to assess beliefs about the causes of poverty, most of the existing measures of perceived causes of poverty lack methodological rigor.” There are three associated problems (Shek, 2002, 2003). First, reliability of the related instruments has been inadequately assessed. This problem is particularly acute in those studies where a belief about the cause of poverty was assessed by a single item (e.g., Hunt, 2002). Second, validity of the existing measures was based generally on face validity only (Shek, 2002, 2003). Third, support for the factorial validity of the existing scales assessing different beliefs about the causes of poverty is weak…Furthermore, to date; no longitudinal studies have been conducted to look at the factorial validity of measures that attempt to assess different beliefs about the causes of poverty (p. 273).

In response to these concerns, Shek (2004; Shek & Ma, 2009) has developed and extensively researched his own scale, the Chinese Perceived Causes of Poverty scale (CPCPS). The CPCPS has been developed with a focus on validity and reliability for use with populations in Hong Kong and with nonuniversity students. Thus, while agreeing with Shek’s assessment of existing instruments, the CPCPS has a different purpose than does the UPPTS.

It was at this point that the authors began developing additional questions that they felt would address the areas of social empathy and commitment to aiding those living in poverty; areas not addressed by Atherton et al. Over time 21 new questions were developed that drew upon a review of the literature (Coryn, 2002; Cozzarelli et al., 2001; Mehrabian & Epstein, l972), the authors’ own experiences in working with the poor and in conducting research with impoverished populations (Blair, 2010; Blair, Taylor, & Rivera, 2009) and feedback from students in classes taught by the authors. The new questions developed for use in the UPPTS build upon the concepts of a lack of social empathy (Segal, 2007) and cognitive distancing (Lott, 2002) as principal reasons why people fail to do more to help the poor via either direct action or support for programs that will aid the poor.

A complete list of the 21 items can be found at: http://sites.niagara.edu/dr-kevin-d-blair/

Theoretical Constructs

The UPPTS consists of three constructs: (1) general attitudes toward those living in poverty, (2) empathy for those living in poverty, and (3) commitment to addressing poverty via direct action or support for programs/services that aid the poor. Further, the authors postulate that general attitudes toward the poor exist in the form of a continuum with the underlying belief about the causes of poverty (i.e., flawed character vs. limited opportunity) laying at the extremes.

In terms of relationships between the constructs, the authors believe that as empathy for the poor increases, so will the respondent’s belief in limited opportunity as being the underlying cause of poverty, and there will be a correlating increase in the respondent’s commitment to addressing poverty. Inversely, as the respondent’s level of empathy decreases, their belief in flawed character as the underlying case of poverty will increase, and their willingness to address poverty will decrease. That is, the more the respondent views poverty as a consequence of poor choices and character flaws, the less empathy they will demonstrate and the less willing they will be to provide assistance.

Method

Procedures in Validation

Hair, Black, Babin, and Anderson (2010) along with Yang (2010) discuss the steps needed to conduct an effective exploratory factor analysis (EFA). Others, such as Rubio, Berg-Weger, Tebb, Lee, and Rauch (2003) discuss the steps needed to validate a new social science measure. Using these methods, and drawing primarily on Hair et al., the authors undertook the following steps to demonstrate the content, criterion, and construct validity of the UPPTS:

Review of the survey items by content experts: The authors have extensive experience in both direct practice and in conducting research with impoverished populations (Blair et al., 2009).

Pilot testing: a draft of the UPPTS was given to 20 undergraduate students who were enrolled in an undergraduate course being taught by one of the authors. Feedback from this pilot testing was used to make changes to the instructions for completing the UPPTS and to clarify a few of the questions.

Collection of data using the 58-item UPPTS was completed. In this part of the process, 301 undergraduates enrolled at a small liberal arts university were surveyed.

Random qualitative validation (RQV) was then conducted via a series of focus groups with students after they had completed the 58-item UPPTS. RQV is, “a method of gathering clarifying qualitative data that improves the validity of the quantitative analysis” (Van Duzer, 2012, p. 1). In essence, RQV involves using a series of focus groups with survey respondents to discuss the respondent’s understanding of both the questions asked in the survey and the validity and accuracy of the available responses. In other words, do the respondents have the same understanding of the questions as intended by the survey’s author/authors and do the available answer choices enable the respondent to respond in a way that accurately reflects their true answer to the question (Van Duzer, 2012).

Results from the 301 completed UPPTS were used to perform EFA using the Statistical Package for the Social Sciences. An α level was set at .05 for all statistical tests.

Sampling Procedures and Participant Characteristics

The UPPTS sample consisted of 301 undergraduate students (see Table 1 for a summary of demographic information). This sample is discussed in depth below, but was found to be representative of the university student body as a whole. The participating students were enrolled in one of the several classes at the university where the authors teach. A majority of the students were enrolled in either a philosophy or religious studies course that is required as part of the university’s general education curriculum. Thus, these students came from each of the university’s four colleges (Arts and Science, Business, Hospitality and Tourism, and Education) and from each of the academic classes (freshmen through seniors) and also drew from traditional 4-year students and transfer students. In addition to these classes, some of these same students were enrolled in one of the three social work courses: Introduction to Social Work, Addressing Poverty (an elective open to all undergraduate students), and Diversity (a course that is required for social work majors and minors, but that is also open to all undergraduate students). Both the Diversity and Poverty class attract a wide range of students who use the course to satisfy general education requirements for a social science and diversity credit. Participants were provided with a consent form explaining that their completion of the UPPTS was voluntary. No students were required to complete the instrument and completion of the UPPTS was in no way connected to the grades students would receive in their respective courses.

Table 1.

Demographic Information for Sample.

	Frequency	%
Gender
Male	113	37.5
Female	184	61.1
Missing	4	1.3
Total	301	100
Age
18–25	288	95.7
26–35	3	1.0
36–50	6	2.0
Missing	4	1.3
Total	301	100
School year
Freshman	40	13.3
Sophomore	66	21.9
Junior	127	42.2
Senior	62	20.6
Missing	6	2.0
Total	301	100
College
Arts and Sciences	168	55.8
Business	48	15.9
Education	21	7.0
Hospitality	43	14.3
Missing	21	7.0
Total	301	100
Ethnicity
White	268	89
African American	9	3.0
Hispanic American	7	2.3
Native American	2	0.7
Other	5	1.7
Missing	10	3.3
Total	301	100
High school type
Public suburban	168	55.8
Public urban	29	9.6
Public rural	39	13
Private catholic	54	18
Private other	6	2.0
Missing	5	1.6
Total	301	100
Religion
Catholic	162	53.8
Protestant	39	13
Jewish	3	1.0
Muslim	1	0.3
Other	72	23.9
Missing	24	8
Total	301	100
Household income
0000–8,000	3	1.0
8,001–15,000	6	2.0
15,001–25,000	13	4.3
25,001–40,000	25	8.3
40,001–60,000	39	13
60,001–80,000	50	16.5
80,000+	95	31.6
Not sure	67	22.3
Missing	3	1.0
Total	301	100

The 301 students who completed the survey demonstrate a profile that is consistent with the overall student body at this small (approximately 3,800 total enrolment), Catholic, suburban, liberal arts university. The typical respondent would most likely be a Caucasian and Catholic female, between the ages of 18 and 25, in her junior year of study, who graduated from a suburban public high school, whose major is found in the College of Arts and Sciences (the largest college on this campus), and whose family owns their home and has a household income over US$60,000 per year. Table 1 provides a summary of the demographic profile of the respondents.

Students who completed the UPPTS were compared to the general university undergraduate student population. Specifically, the authors performed a series of chi-squared tests for goodness of fit. For gender, our sample showed 60% female to 40% males that matched the university population proportions of males and females. Also for religious affiliation, the percentage of catholic, protestant, and other religious backgrounds was found to be reasonably accurate (χ² = 4.91, df = 2, p = .09). However, the distribution of ages was statistically different from the general University population (χ² = 26.92, df = 1, p < .001, ω = .30). Our sample had 97% of respondents between the ages of 18 and 24 as opposed to 88% in the general university population. Similarly, the distribution of race/ethnicity was also found to be statistically different from the population (χ² = 60.54, df = 2, p < .001; ω = .45). This was due to a larger percentage of Whites in the sample, 92%, compared to the general university population of 72%.

Distribution by college in the sample group also differed significantly from the general university population (χ² = 12.33, df = 3, p = .006; ω = .21). This was due to a slightly larger percentage of students from the college of arts and sciences and a smaller percentage of students sampled from the college of education. Because of these differences, some care needs to be taken when generalizing these results to the entire university undergraduate population. However, since the sample size is over 300, the statistical comparisons tend to be very sensitive to small deviations. Since most of the university students are White and between the ages of 18 and 24, the authors feel the sample provides a solid representation of the students who attend this university.

Independent one-way analysis of variance (ANOVA) tests were performed on the sample of 301 respondents to test the homogeneity of the various demographics. The total score on all the questions of the UPPTS after some of the questions were reverse scored was used as the dependent variable. Since all the questions had scores from 1 to 5, we transformed the total score to be $S = \frac{100 (\sum x - N)}{4 N}$ where N is the number of questions. This provides a score between 0 and 100 where the lower the score, the more compassionate the student feels toward the poor. Whereas the higher the score, the more the student believes that being poor is the result of a flaw in the character of people who live in poverty. The one-way ANOVA tests came out insignificant for religion, F(2, 246) = 0.12, p = .89, class year, F(3, 260) = 0.44, p = .73, race/ethnicity, F(1, 259) = 0.20, p = .67, and type of high school, F(3, 244) = 1.43, p = .24. Thus, the mean total score over all the questions may be the same across various religions, class years, races, and types of high school the student attended. In other words, religious affiliation (catholic, protestant, other), class year (freshman, sophomore, junior, or senior), and ethnicity/racial identification showed no effect on the respondents’ answers to the UPPTS questions.

However, gender found a difference in mean total score on the UPPTS, t(264) = 4.25, p < .001, d = .54. Men had significantly higher total UPPTS score (M = 43.55, SD = 13.35) than women (M = 36.30, SD = 13.50). This means that female students are more likely to be empathic toward the poor, whereas male students are more likely to view the poor negatively and as having a flaw in their character that leads them to be in poverty. There was also a difference in mean total UPPTS score across the various colleges, F(3, 242) = 9.49, df = 3, p < .001, η² = .06, but the effect size was small.

Running Bonferroni post hoc tests, controlling for simultaneous tests, it was found that the students whose major is within the college of arts and sciences had a lower mean score (M = 35.72, SD = 5.72) than the colleges of business (M = 45.42, SD = 14.09), education (M = 46.19, SD = 11.07), and hospitality (M = 43.24, SD = 11.41). Thus, students who major in the arts and sciences tend to be more empathic to the poor than students enrolled in other colleges across the university.

An analysis of the relationship between family income and UPPTS score was also revealing. Initially, a scatter plot showing the relationship between income and total UPPTS score after the average income in each bracket was recorded. From the scatter plot, it appeared that there was a lot of heteroscedasticity in the data. Students whose income was low tended to score low on the UPPTS, meaning that if they grew up poor, they tended to be very compassionate toward the poor. Students whose income was high, however, tended to have a greater variation in UPPTS scores. Since the heteroscedasticity violated the assumptions for linear regression, a bivariate Spearman’s test was performed. This test showed a positive correlation (ρ = .20, df = 207, p = .004) between income and UPPTS scores. In other words, the higher a family’s income, the more likely it is that the respondent holds a negative view of those who live in poverty. The lower a family’s income, the more likely it is that the respondent holds a positive (more empathetic/sympathetic) view of those who live in poverty.

In general, these results are consistent with the authors’ expectations and suggest that the UPPTS is yielding useful and accurate information about the attitudes of respondents toward those who live in poverty. Further, limitations created by the sample group are discussed later, but the sample does seem representative of the university undergraduate population as a whole and thus while wanting to be careful, it is highly likely that these results can be generalized to undergraduate students at similar universities.

RQV

A total of five focus groups involving 24 students who had completed the UPPTS were held within a few days immediately following the initial survey. Participants were volunteers from the courses being taught by the researchers. Participants were not offered any remuneration for their participation. Focus group participants were exclusively from the social work courses and were generally students who were more interested in the topic of poverty and/or the research process. The typical focus group participant matched the typical survey respondent: female, Catholic, Caucasian, suburban public high school, and majoring in the college of arts and sciences. Twenty of the focus group participants were traditional students between the ages of 18 and 22, while the remaining 4 were nontraditional. Of the nontraditional students, three were female and all four were in their mid- to late 30s. Of the 20 traditional students, 12 were female. All but two of the students were in either their junior or senior year of college. Despite the demographic differences between the focus group participants and the general university population, the researchers felt that the information collected was very helpful and the information gained provided valid insight into the UPPTS that could be used to supplement the EFA.

The focus groups were led by three of the authors who have qualitative research experience and who have led many focus groups. Each focus group had four to six participants and focused on 12–15 of the initial 58 UPPTS questions and explored with participants their understanding of the meaning of the questions and the extent to which it matched the author’s intent and the extent to which the Likert-type scale answer choices provided the opportunity for the respondent to accurately answer the questions.

Respondents and Sampling Adequacy for EFA

In terms of adequacy for conducting an EFA, Hair et al. (2010, p. 105) advise that four criteria need to be met:

“A strong conceptual foundation needs to support the assumption that a structure does exist before the factor analysis is performed.” The conceptual foundation for the UPPTS and for the assumption of underlying structure has been established in the literature review and theoretical sections.

“Measure of sampling adequacy (MSA) values must exceed .50…” The score for the Kaiser–Meyer–Olkin measure of sampling adequacy was .86 indicating that the sample size was adequate for EFA.

“A statistically significant Barlett’s test of sphericity with a chi-squared value of 4075.51 and a p-value < .001 indicates that sufficient correlations exist among the variables to proceed [with an exploratory factor analysis].” Barlett’s test of sphericity was statistically significant.

Finally, the 301 respondents provide an initial ratio of just over five observations per variable for the 58 questions. This meets the criteria for observations per variable for effective factor analysis (Hair, Black, Babin, & Anderson, 2010, p. 102).

Results

EFA

Principal factor extraction with promax rotation was performed on 58 items from the sample of 301 subjects. Listwise deletion was used to handle the 30 subjects with missing data bringing the total sample size at 271. Prior to deleting the data, chi-square tests were performed on all of the demographics to determine whether the missing data were missing at random. The p values from those tests ranged from .42 to .83 indicating that the data were missing at random. Principal component analysis was used prior to principal factors extraction to estimate the number of factors. Using the Kaiser–Guttman criterion of Eigenvalues greater than 1 gave us an upper bound on the number of factors to retain at 16 factors. The scree plot appeared to contain a break around three factors and another break around eight factors. We also conducted a parallel analysis that generates multiple random data sets with the same number of cases and variables as our data set. Comparing the Eigenvalues in our sample to the Eigenvalues in the simulated data set indicates that a seven-factor solution would be reasonable. Since the parallel analysis has a solution near the second break in the scree plot, and is the gold standard in EFA, we decided to run a seven-factor model (Hair et al., 2010).

The seven-factor model explained 50.44% of the variance. Loadings of variables on factors, communalities, and percentage of variance explained are all given in Table 2. Variables are ordered and grouped by size of loading to facilitate interpretation. With a cutoff of .35 for inclusion of a variable in interpretation of a factor, we retained 42 of the 58 items. This model contains simple structure with no variable loading on more than one factor. These seven factors were then further analyzed for their reliability.

Table 2.

Factor Loadings, Communalities (h ²), and Percentage of Variance Explained for Principal Factor Extraction With Promax Rotation.

Question	F ₁	F ₂	F ₃	F ₄	F ₅	F ₆	F ₇	h ²
There is a lot of fraud among welfare recipients	.81							.55
Welfare makes people lazy	.60							.55
Poor people use food stamps wisely (r)	.60							.31
The government spends too much money on poverty programs	.56							.54
Benefits for poor people consume a major part of the federal budget	.56							.44
Welfare mothers have babies to get more money	.55							.41
Welfare recipients should be able to spend their money as they choose (r)	.47							.18
I would support a program that would result in higher taxes to support programs for poor people (r)	.47							.43
People in welfare should be made to work for their benefits	.46							.23
A person receiving welfare should not have a nicer car than I do	.44							.21
Some poor people live better than I do considering all their benefits	.41							.20
An abled-bodied person using food stamps is ripping off the system	.40							.29
Poor people act differently		.81						.45
I believe poor people have a different Set of values than do other people		.63						.36
Poor people are different from the rest of society		.60						.27
Poor people generally have lower intelligence than nonpoor people		.55						.34
Most poor people are dirty		.49						.34
Most people who are poor waste a lot of their time		.42						.40
Being poor is a choice		.38						.36
I believe poor people create their own difficulties		.38						.45
Individuals should do more to help the poor (r)			.76					.68
Businesses should do more to help the poor (r)			.60					.54
Society has the responsibility to help poor people (r)			.50					.44
Charities should do more to help the poor (r)			.49					.37
I feel that I could personally make a difference in the lives of the poor (r)			.49					.41
It upsets me to know that people are poor (r)			.36					.40
The poor have the same opportunities for success as everyone else				.82				.49
If poor people worked harder they could escape poverty				.62				.50
Any person can get ahead in this country				.61				.37
The poor face challenges that are the same as everyone else				.59				.28
Unemployed poor people could find jobs if they tried harder				.44				.50
Poor people are satisfied receiving welfare				.40				.31
Everyone regardless of circumstances should have a place to live (r)					.89			.70
Everyone regardless of circumstances should have health care (r)					.79			.64
Everyone regardless of circumstances should have enough food to eat (r)					.74			.57
Lack of social support is a major challenge for the poor (r)						.66		.43
Lack of Child Care is a major challenge for the poor (r)						.62		.46
Lack of education is a major challenge for the poor (r)						.62		.36
Lack of Transportation is a major challenge for the poor (r)						.51		.37
I feel that I know what it is like to be poor (r)							.67	.48
I feel that I have enough direct exposure to the poor (r)							.66	.47
I feel that I understand why someone may be poor (r)							.35	.23
Percentage of variance explained		21.54	6.78	5.58	4.90	4.05	4.00	3.59
Cronbach’s α		.84	.77	.82	.72	.83	.71	.56

Note. Factor scores of greater than .35 are reported, including secondary factor loadings.

(r) indicates an item that was reverse scored.

Factor labels: F ₁: Welfare attitudes (WA), F ₂: Poor are different. (PD), F ₃: Do more (DM), F ₄: Equal Opportunity (EO), F ₅:Fundamental Rights (FR), F ₆: Lack of Resources (LR), F ₇: Social Empathy (SE).

Tests for internal reliability are given for all seven of the factors. The first six factors have Cronbach’s α values above .7, the acceptable standard. The last factor, social empathy, has a Cronbach’s α score at .56. We decided not to retain the last factor due to its low reliability. While the authors were somewhat disappointed that social empathy could not be retained as a factor, many of the questions that the authors had developed regarding social empathy did load onto the Poor Are Different (PD, F2) and Do More (DM, F3) factors. Each factor is discussed in more depth in the next section.

As recommended by Hair et al. (2010) and others, Promax rotation was used because there is correlation between the factors. The correlation matrix is given in Table 3. The correlations between the first six factors ranged from .17 to .54. All of the correlations for the first six factors were significantly different from zero with p values < .01. The social empathy factor seems uncorrelated with the other factors.

Table 3.

Correlation of Retained Factors.

	WA	PD	DM	EO	FR	LR	SE
WA	1.00
PD	.51***	1.00
DM	.41***	.45***	1.00
EO	.54***	.41***	.35***	1.00
FR	.30***	.37***	.40***	.21***	1.00
LR	.25***	.18***	.41***	.26***	.17***	1.00
SE	.08	−.03	.22***	0.00	.08	.05	1.00

Note. WA = Welfare Attitude; PD = Poor Are Different; DM = Do More; EO = Equal Opportunity; FR = Fundamental Rights; LR = Lack of Resources; SE = Social Empathy.

*p < .05. **p < .01. ***p < .001.

In summary, the six retained factors are “Attitudes Towards Welfare,” “Do more,” “Poor are different,” “Equal Opportunity,” “Fundamental Rights,” and “Lack of Resources.”

Discussion of Factors, Factor Labels, and Relationships Between the Factors

In general, the authors were pleased to see the factors derived from the EFA, as they conformed to the underlying constructs and theoretical positions. In interpreting the factors and their meanings, the authors drew on their own experience and on information that had been obtained via follow-up focus groups with students. Each factor is discussed in depth in the following sections. We include some limited discussion of the actual results to help illustrate each factor.

The first factor is labeled Welfare Attitude (WA). Each of the questions in this factor relates directly to welfare programs such as the Supplemental Nutrition Assistance Program, which is more generally known as Food Stamps. This is the largest factor with 12 questions and it explains 21.54% of the variance (see Table 2). Focus group discussions indicated that students in this sample generally hold negative attitudes toward welfare programs and by inference, primarily negative attitudes toward persons who accept welfare assistance. Students with the most negative attitude toward welfare also tended to hold very negative attitudes toward the poor as indicated by more negative scores on the other factors. Also focus group discussions indicated that this factor also relates to the respondent’s level of empathy to the poor, that is, those with more empathy felt it was okay to accept help. Those with less empathy felt accepting help was a clearer sign of flawed character and not a reasonable response to the situation. Focus groups also showed that these students differentiate welfare assistance from other supports such as federally subsidized student loans (which the majority of students in the sample receive), work study programs, and so forth.

Questions in Factor 2 all relate to the respondent’s sense of the poor as being significantly different or “other” and, as can been seen from questions such as “Most poor people are dirty,” there is a negative assessment associated with the otherness. Focus group discussions indicated that while some students saw such differences as cultural and neutral, most viewed the differences as negative and some viewed the poor as clearly inferior. This sense of the poor as “other” relates directly to Lott’s (2002) discussion of cognitive distancing and to Segal’s (2007) position that a lack of social empathy permeates the view of the poor. While this sample is too small to generalize, the authors hypothesize this sense of the poor as other will be typical of undergraduates at similar institutions throughout the United States.

Factor 3 has been labeled Need to Do More (DM) and the six questions all relate to the need for institutions and individuals do more to provide assistance. Fully 63% of respondents in our sample do not see a need to provide more assistance to those in poverty. Here again, the authors feel that empathy plays a large role in the respondent’s thinking. Focus group discussion indicated that students in this sample generally think: “the poor already get enough assistance, society has done its part, the rest is up to the poor.” This conviction appeared to translate to a firm belief that if the poor worked harder, they could overcome their poverty.

Factor 4 has been labeled Equal Opportunity (EO). Focus group discussions indicate that many students are conflicted about whether the poor have EO. This conflict also appears in the discussion of Fundamental Rights (FR, F5) and Lack of Resources (LR, F6). As discussed in the focus groups, respondents want to believe in the ability of individuals to overcome all obstacles; that is, anyone can succeed if they try hard enough on one hand; but, on the other hand, respondents acknowledge that access to resources and social supports are vital.

Factor 5 has been labeled Fundamental Rights (FR). Students at this small university hold very negative views of basic rights to food, shelter, and health care. Fully 62% of the respondents tended to disagree with a basic right to food, shelter, and health care. Students tend to believe that a person must work and earn these basics.

Factor 6 has been labeled Lack of Resources (LR). These four questions deal with the respondents sense of whether the poor have the resources or have access to the resources needed to change their situation. Here again, focus group discussions showed that respondents are conflicted in this area. Many of the students are very willing to acknowledge that the poor have limited resources and face difficult challenges, but the majority seemed to believe that “when the going gets tough the tough get going.” A smaller group, again showing what the authors interpret as higher levels of empathy, felt that living in poverty, “can beat you down,” one student put it.

Finally, Factor 7 consists of three questions that were part of a larger group added by the authors in an effort to assess the respondents’ level of social empathy. The authors had hoped that this factor would stand alone, but the questions tended to load onto either DM or PD, while a few others failed to load and were eliminated. The authors’ belief is that the factors DM and PD are related to the respondents’ level of empathy and cognitive distancing. That is, a willingness to do more to help the poor and viewing the poor as similar rather than different indicates less cognitive distancing (Lott, 2002) and thus more empathy (Segal, 2007; Weiss, 2006).

Discussion and Applications to Social Work

The EFA in combination with the focus groups shows that the UPPTS is providing accurate and useful information regarding the respondent’s attitudes and perceptions about those living in poverty. As currently arranged, the lower a respondent’s cumulative score on the six factors, the more favorable their perception of the poor and, as confirmed during the focus groups, the more likely the respondent is to view the underlying cause of poverty as being limited resopurces and opportunities. In contrast, the higher the respondent’s cumulative score, the less favorable is their perception of those living in poverty and the more likely the respondent is to view the underlying cause of poverty to be some form of character flaw and poor decision making by those living in poverty. This is vital in light of research that indicates students who perceive character flaws as the cause of poverty are less interested in working with poor people (Weiss, 2006).

For educators, the UPPTS can provide important baseline information regarding the attitudes of a class toward the poor and thus indicate where the instructor needs to most focus. Table 4 provides five case studies to help illustrate the usefulness of the UPPTS. The five cases presented in the table represent the five quartiles of the distribution. Case 1 holds the lowest (most favorable) total score for the six factors at 47. Case 5 holds the highest (most negative) score at 165. While these cumulative scores are interesting, the subscale scores are more revealing and likely more useful for an instructor who wants to fine tune their teaching, reading, and experiential assignments to best address gaps and misperceptions of their students.

Table 4.

Case Examples Illustrating the Factors.

Survey Questions	Case 1	Case 2	Case 3	Case 4	Case 5
Factor 1: welfare attitude
There is a lot of fraud among welfare recipients	1	3	3	5	5
Welfare makes people lazy	1	2	1	1	4
Poor people use food stamps wisely (r)	3	2	4	3	3
Benefits for poor people consume a major part of the federal budget	1	3	3	5	2
The government spends too much money on poverty programs	1	2	3	1	5
Welfare mothers have babies to get more money	1	3	3	5	5
Welfare recipients should be able to spend their money as they please (r)	1	3	3	5	4
I would support a program that would result in higher taxes to support programs for poor people (r)	1	3	3	3	4
People in welfare should be made to work for their benefits	5	3	4	5	5
A person receiving welfare should not have a nicer car than I do	1	3	5	5	5
Some poor people live better than I do considering all their benefits	1	2	2	3	2
An able-bodied person using food stamps in ripping off the system	1	2	3	5	5
Subtotal	18	31	37	46	49
Factor 2: poor are different
Poor people act different	1	2	3	5	5
I believe poor people have a different set of values from other people	1	3	3	3	5
Poor people are different from the rest of society	1	2	3	1	3
Poor people generally have lower intelligence than nonpoor people	1	4	4	5	5
Most poor people are dirty	3	2	3	3	4
Most people who are poor waste a lot of their time	1	2	2	3	5
Being poor is a choice	1	1	3	1	5
I believe poor people create their own difficulties	1	2	3	3	5
Subtotal	10	18	24	24	37
Factor 3: Do More
Individuals should do more to help the poor (r)	1	2	3	1	4
Businesses should do more to help the poor (r)	1	2	3	1	4
Society has the responsibility to help poor people (r)	1	2	1	1	5
Charities should do more to help the poor (r)	1	2	3	3	4
I feel I could personally make a difference in the lives of the poor (r)	1	3	3	1	4
It upsets me to know that people are poor (r)	1	2	2	3	4
Subtotal	6	13	15	10	25
Factor 4: Equal Opportunity
The poor have the same opportunities for success as everyone else	1	2	3	1	5
If poor people worked harder they could escape poverty	1	1	3	5	5
Any person can get ahead in this country	1	2	2	5	5
The poor face challenges that are the same as everyone else	1	2	2	1	1
Unemployed poor people could find jobs if they tried harder	1	2	2	5	5
Poor people are satisfied receiving welfare.	1	2	1	3	5
Subtotal	6	11	13	20	26
Factor 5: Fundamental Rights
Everyone regardless of circumstances should have a place to live (r)	1	2	1	1	4
Everyone regardless of circumstances should have health care (r)	1	2	1	1	4
Everyone regardless of circumstances should have enough food to eat (r)	1	2	1	1	4
Subtotal	3	6	3	3	12
Factor 6: Lack of Resources
Lack of social support is a major challenge for the poor (r)	1	1	2	5	4
Lack of Child Care is a major challenge for the poor (r)	1	1	2	5	4
Lack of education is a major challenge for the poor (r)	1	1	2	5	4
Lake of Transportation is a major challenge for the poor (r)	1	2	2	1	4
Subtotal	4	5	8	16	16
Composite total	47	84	100	119	165

Note. Survey responses: 1 = strongly agree; 2 = agree; 3 = neutral/no opinion; 4 = disagree; 5 = strongly agree.

Items designated (r) are reversed scored. A lower score shows a more positive and empathic view of those living in poverty, while a higher score shows a less positive and less empathic view.

Basic statistics on the composite total: M = 100.9, Median = 100, SD = 24.5.

The lowest possible score is a 39. Lowest actual score was a 46. Highest possible score is 195. Actual highest score was 165. The five cases represent the five quartiles of the distribution.

Instructors can use the survey to obtain a baseline of the attitude toward the poor by a group of students. Following the end of the course, the survey can be conducted a second time to provide a general assessment of changes in attitude that have occurred.

The subscales can also be used to focus the instructor’s attention on areas that seem to need the most attention, those showing the greatest potential for misunderstanding of those that live in poverty.

Looking at the first subscale (Factor 1, WA), the case examples show that overall students at this small, suburban institution hold a fairly negative view of welfare, government assistance, and of welfare recipients. The lowest possible (most favorable view of welfare) score would be 12. None of the 301 respondents scored this low score. The lowest recorded score for WA was 18 (see Table 4). The highest possible score is 60 and the midpoint would be 30. Case 1, the respondent with the lowest total score does have a positive view of welfare assistance, but still feels (very strongly) that welfare recipients should be made to work for their benefits, indicating a lack of understanding of the requirements for programs such as Temporary Assistance to Needy Families. In contrast, Case 5 shows a WA score of 49 that is 11 points below the highest negative score of 60. This suggests that while this respondent has a fairly negative view of welfare, they may be open to learning more and to discussing the pros and cons of assistance programs.

Need to Do More (DM) has six questions that relate to the need for entities other than government to do more to provide assistance. The case examples show that, in general, these students do not see a need for more assistance. The highest possible score is 30, which is also the mode score for this factor. A score of 30 was recorded by 53 respondents—almost 18% of the 301 respondents. Of the respondents, 45% in this sample score 25 or above. Taken together, fully 63% of respondents do not see a need to provide more assistance to those in poverty. Given the relationship of this factor to the respondents’ level of empathy for those living in poverty, it is clear that this is an area that needs significant attention. In fact, this information has been used by one of the authors to focus more readings and assignments in the course on poverty on assistance programs and in particular the challenges faced by local food pantries, soup kitchens, and other privately funded programs to meeting the actual needs of the impoverished community.

Questions in Factor 3 all relate to the respondent’s sense of the poor as being significantly different or “other” and, as can been seen from questions such as “Most poor people are dirty,” there is a negative assessment associated with the otherness. Focus group discussions indicated that while some students saw such differences as cultural and neutral, most viewed the differences as negative and some viewed the poor as clearly inferior. There is a wide range of responses recorded for this factor. Of the respondents, 50% scored 12 (six is the lowest possible, most favorable score) or below and 50% scored 13 or above. Of the respondents, 31 scored a 6, the lowest possible score while 3 respondents (including Case Example 5) scored a 25 or higher (30 is the highest, most negative possible score). This sense of the poor as “other” relates directly to Lott’s (2002) discussion of cognitive distancing and to Segal’s (2007) position that a lack of social empathy permeates the view of the poor. While this sample is too small to generalize, the authors hypothesize this sense of the poor as other will be typical of most undergraduates throughout the United States and particularly those attending smaller, liberal arts colleges.

Factor 5 has been labeled Fundamental Rights (FR). Students at this small university hold very negative views of basic rights to food, shelter, and health care. Less than 10% of the respondents scored below 9. Fully 62% of the respondents scored 15, the highest and most negative possible score. Even if students held a more favorable view of the poor, they tended to disagree with a basic right to food, shelter, and health care. Students tend to belief that a person must work and earn these basics.

Factors 4 and 6, respectively, identified as Belief in EO, LR appear to be closely related. Focus group discussions indicate that many students are conflicted in these areas. They want to believe in the ability of individuals to overcome all obstacles; that is, anyone can succeed if they try hard enough on the one hand; but, on the other hand, know that access to resources and social supports are vital.

All in all, the authors felt that the results were consistent with both Lott (2002) and Lane’s (2001) arguments that most people in the United States (in this case, undergraduate students) hold strong opinions about those living in poverty without having much direct experience on which to base their beliefs. Focus groups confirmed that students hold preexisting opinions about the poor and the reasons why people are poor while readily admitting their limited exposure to poverty. Students with the strongest opinions (positive or negative) about those living in poverty also seemed to be the ones who felt the least need for more contact and direct experience with poverty. In fact, the authors felt that some students were concerned that their firmly held beliefs (again positive or negative) would be undermined by more direct experience with poverty. Following opportunities to directly interact with poor people, students may find their preconceptions about poverty are incompatible with the realities of poverty (Zygmunt-Fillwalk, 2009).

Implications are that instructors should be including class activities that provide direct contact and exposure to people who are living in poverty, as well as in class exercises (such as simulations) that may help increase empathy and understanding of what it is like to live in poverty. We are encouraged by evidence that participation in a poverty simulation facilitates more positive attitudes toward the poor (Patterson & Hulton, 2012) and results in more empathy and compassion for those who are poor (Nickols & Nielsen, 2011). Evidence suggests that community-based experiences involving interactions with people living in poverty can foster positive changes in beliefs and attitudes about poverty (Gumpert & Kraybill-Greggo, 2005; Proctor et al., 2010). Such activities in conjunction with a pedagogical emphasis on structural determinants of poverty could have a powerful sway on students’ attitudes (Weaver & Yun, 2010). The UPPTS can be a useful tool, helping instructors track the effectiveness of class assignments and impact of course content.

The main limitation derives from the sample, which is drawn from one relatively small, suburban university. While this group of undergraduates is likely to be similar to the students found at many comparable universities, it is equally likely that students from larger, public, urban institutions will be quite different and their perceptions of those living in poverty may also be quite different. Further, this sample draws largely on a sample group of undergraduates with little experience of being impoverished.

Further, as noted by Hair et al. (2010, p. 139), “Validation of any factor analysis result is essential…;” thus, the results presented should be considered preliminary pending further validation. The authors will soon be collecting data from a larger, urban university. This should result in not only a second set data that can be compared to this first set but should push the total number of respondents to well over 1,000 which will provide well over 10 observations per item. In addition to comparisons between the two sets of data, the authors intend to use confirmatory factor analysis to further validate the UPPTS.

Educators and others interested in learning more about their students’ perceptions of those living in poverty can use the UPPTS to obtain valuable information and to track changes in perceptions over time. The survey provides important improvements over the original Atherton et al. instrument giving a more in-depth understanding of the respondent’s attitudes and perceptions regarding the poor and in particular significant information regarding the respondent’s beliefs about the underlying causes of poverty, their belief/attitudes about welfare and government interventions, and to some extent, a sense of the level of empathy the respondent has for those living in poverty. While we anticipate making enhancement to the UPPTS, these 39 questions and six factors provide users with a good baseline regarding their student’s perceptions of the poor and the ability to track the impact of their coursework and related activities on these perceptions.

Table 5.

Instructions for Administering and Scoring the UPPTS.

The UPPTS is designed to assess a respondent’s overall attitude toward those that are living in poverty. Questions that have an * are reverse scored (i.e., 1 becomes 5 and 5 becomes 1 and so forth). The asterisk should be used only for scoring and should removed before administering the UPPTS.
The higher a respondent’s final total score, the more negatively they view those living in poverty, the higher score also indicates a lower level of overall empathy for those living in poverty, and it is more likely that the respondent views those living in poverty as having some type of flawed character and that the underlying cause of poverty is the flawed character of those in poverty.
The lower a respondent’s final score total score, the more likely they are to have a positive, more empathic view of those in poverty and the more likely they are to view the underlying cause of poverty as lack of resources and opportunity.
Most respondents will find themselves somewhere in the middle of these two poles.
Respondents’ scores can be further broken down into six subcategories: Attitudes about welfare programs and those who are willing to accept welfare supports (WA; Questions 1–12). Poor are different from the nonpoor (PD; Questions 13–20) shows the extent to which the respondent thinks of the poor as some type of “other” and not as one of “us.” PD is significant in the respondent’s willingness to help those living in poverty either through direct action or via support for programs and services that aid the poor. Do More (DM; Questions 21–26) illustrate the respondent’s willingness to do more via either direct action or via support for programs and services. Belief in equal opportunity (EQ; Questions 27–32) indicates the respondent’s belief in whether everyone has a good chance to succeed. A higher score indicates a belief in overall equal opportunity and thus also indicates a belief that those in poverty have only themselves to blame for their situation. A belief in basic rights to food, clothing, shelter, and so forth (FR; Questions 33–35) is assessed by this subscale. A lower score indicates a belief that access to these basics for survival is a fundamental right and not conditional or based on concepts of deserving/undeserving. The last subscale (LR, Questions 36–39) reflects the respondent’s attitude about resources and their importance in whether or not someone lives in poverty. These question also reflect somewhat on the level of empathy the respondent has for those living in poverty and their understanding of the challenges faced by those in poverty.
UPPTS Scoring Form (items marked with * are reverse scored)
Question Number	Score	Question Number	Score
1.		13.
2.		14.
3.*		15.
4.		16.
5.		17.
6.		18.
7.		19.
8.*		20.
9.		PD Total
10.
11.
12.
WA Total
Question Number	Score	Question Number	Score
21.*		27.
22.*		28.
23.*		29.
24.*		30.
25.*		31.
26.*		32.
DM Total	EQ Total
Question Number	Score	Question Number	Score
33.*		36.*
34.*		37.*
35.*		38.*
FR Total		39.*
		LR Total
WA Total:_____
PD Total: _____
DM Total _____
EQ Total _____
FR Total _____
LR Total _____
Overall Score: _____
Instructions:
Circle the answer that best matches the extent to which you agree or disagree with each statement.
			5 4 3 2 1
1. There is a lot of fraud among welfare recipients			SA A N D SD
2. Welfare makes people lazy			SA A N D SD
3. Poor people use food stamps wisely*			SA A N D SD
4. The government spends too much money on poverty programs			SA A N D SD
5. Benefits for poor people consume a major part of the federal budget			SA A N D SD
6. Welfare mothers have babies to get more money			SA A N D SD
7. Welfare recipients should be able to spend their money as they choose (r).			SA A N D SD
8. I would support a program that would result in higher taxes to support programs for poor people *			SA A N D SD
9. People in welfare should be made to work for their benefits			SA A N D SD
10. A person receiving welfare should not have a nicer car than I do			SA A N D SD
11. Some poor people live better than I do considering all their benefits			SA A N D SD
12. An abled-bodied person using food stamps is ripping off the system			SA A N D SD
13. Poor people act differently			SA A N D SD
14. I believe poor people have a different set of values than do other people			SA A N D SD
15. Poor people are different from the rest of society			SA A N D SD
16. Poor people generally have lower intelligence than nonpoor people			SA A N D SD
17. Most poor people are dirty			SA A N D SD
18. Most people who are poor waste a lot of their time			SA A N D SD
19. Being poor is a choice			SA A N D SD
20. I believe poor people create their own difficulties			SA A N D SD
21. Individuals should do more to help the poor*			SA A N D SD
22. Businesses should do more to help the poor*			SA A N D SD
23. Society has the responsibility to help poor people*			SA A N D SD
24. Charities should do more to help the poor*			SA A N D SD
25. I feel that I could personally make a difference in the lives of the poor*			SA A N D SD
26. It upsets me to know that people are poor*			SA A N D SD
27. The poor have the same opportunities for success as everyone else			SA A N D SD
28. If poor people worked harder they could escape poverty			SA A N D SD
29. Any person can get ahead in this country			SA A N D SD
30. The poor face challenges that are the same as everyone else			SA A N D SD
31. Unemployed poor people could find jobs if they tried harder.			SA A N D SD
32. Poor people are satisfied receiving welfare			SA A N D SD
33. Everyone regardless of circumstances should have a place to live*			SA A N D SD
34. Everyone regardless of circumstances should have health care*			SA A N D SD
35. Everyone regardless of circumstances should have enough food to eat*			SA A N D SD
36. Lack of social support is a major challenge for the poor*			SA A N D SD
37. Lack of Child Care is a major challenge for the poor*			SA A N D SD
38. Lack of education is a major challenge for the poor*			SA A N D SD
39. Lack of Transportation is a major challenge for the poor*			SA A N D SD

The final version of the UPPTS for use by readers can be found in Table 5 along with instructions for administering the UPPTS and scoring. We also suggest that users of the UPPTS add a section of demographic questions (e.g., year in school, age, gender, major, college, number of poverty related courses completed, self-described socioeconomic status, and so forth), depending on the specific needs and situation. A suggested set of demographic questions can be found at http://sites.niagara.edu/dr-kevin-d-blair/.

Footnotes

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Abouchedid

Nasser

(2001). Poverty attitudes and their determinants in Lebanon’s plural society. Journal of Economic Psychology, 22, 271–282.

Atherton

Gemmel

Haagenstad

Holt

Jensen

O’Hara

Rehner

(1993). Measuring attitudes toward poverty: A new scale. Social Work Research & Abstracts, 29, 28–31.

Austin

(2007). Understanding poverty: Multiple social science perspectives. Journal of Human Behavior in the Social Environment, 16, 1–4.

Blair

(2010). Teaching undergraduates about poverty: An overview. In P. Kriese & R. E. Osborne (Eds.), Social justice, poverty and race: Normative and empirical points of view (109–123) New York: Rodopi Press.

Blair

Taylor

Rivera

(2009). Strengths and stressors in a population of kinship caregivers: Implications for caseload management and administration. Families in Society, 90, 431–438.

Buss

(2010). Have the poor gotten poorer?: The American experience from 1987 to 2007. Journal of Poverty, 14, 183–196.

Clark

(2007). Social work students’ perceptions of poverty. Journal of Human Behavior in the Social Environment, 16, 149–166.

Coryn

C. L. S.

(2002). Antecedents of attitudes toward the poor. Indiana University South Bend Undergraduate Research Journal, 5, 13–19.

Cozzarelli

Wilkinson

A. V.

Tagler

M. J.

(2001). Attitude toward the poor and attributions for poverty. Journal of Social Issues, 57, 207–227.

10.

Gumpert

Kraybill-Greggo

J. W.

(2005). Affecting attitudes toward the poor through group process: The alternative break service trip. Journal of Higher Education Outreach and Engagement, 10, 7–22.

11.

Hair

Jr Black

Babin

Anderson

(2010). Multivariate data analysis (7th ed.). Upper saddle River, NJ: Prentice Hall.

12.

Harrington

(1962). The other America: Poverty in the United States. New York, NY: Macmillan.

13.

Hunt

M. O.

(2002). Religion, race/ethnicity, and beliefs about poverty. Social Science Quarterly, 83, 810–831.

14.

Lane

(2001). Self-reliance and empathy: The enemies of poverty and the poor. Political Psychology, 22, 473–492.

15.

Lott

(2002). Cognitive and behavioral distancing from the poor. American Psychologist, 57, 100–110.

16.

Mehrabian

Epstein

(l972). A measure of emotional empathy. Journal of Personality, 40, 525–543.

17.

Nasser

(2007). Does subjective class predict the causal attribution for poverty? Journal of Social Sciences, 3, 197–201.

18.

Nickols

Nielsen

. (2011). “So many people are struggling”: Developing social empathy through a poverty simulation. Journal of Poverty, 15, 22–42.

19.

Patterson

Hulton

(2012). Enhancing nursing students’ understanding of poverty through simulation. Public Health Nursing, 29, 143–151.

20.

Proctor

Lake

Jewell

Racine

D’Eon

Reeder

(2010). Influencing student beliefs about poverty and health through interpersonal community-based educational experiences. Journal of Research in Interprofessional Practice and Education, 1, 145–158.

21.

Rehner

Ishee

Salloum

Velasques

(1997). Mississippi social workers’ attitudes toward poverty and the poor. Journal of Social Work Education, 33, 131–142.

22.

Reutter

Neufeld

Harrison

(1999). Public perceptions of the relationship between poverty and health. Canadian Journal of Public Health, 90, 13–18.

23.

Reutter

Sword

Meagher-Stewart

Rideout

(2004). Nursing students’ beliefs about poverty and health. Journal of Advanced Nursing, 48, 299–309.

24.

Rubio

Berg-Weger

Tebb

Lee

E. S.

Rauch

(2003). Objectifying content validity: Conducting a content validity study in social work research. Social Work Research, 27, 94–103.

25.

Segal

(2007). Social empathy: A new paradigm to address poverty. Journal of Poverty, 11, 65–81.

26.

Shek

D. T. L.

(2002). Chinese adolescents’ explanations of poverty: The perceived causes of poverty scale. Adolescence, 37, 789–803.

27.

Shek

D. T. L.

(2003). Chinese people’s explanations of poverty: The perceived causes of poverty scale. Research on Social Work Practice, 13, 622–640.

28.

Shek

D. T. L.

(2004). Beliefs about causes of poverty and psychological well-being of parents and adolescents experiencing economic disadvantage in Hong Kong. American Journal of Family Therapy, 32, 239–254.

29.

Shek

D. T. L.

(2009). Dimensionality of the Chinese perceived causes of poverty scale: Findings based on confirmatory factor analysis. Social Indicators Research, 90, 155–164.

30.

Smith

(1943). A tree grows in Brooklyn. New York, NY: Perennial Classics 1st Perennial Classics Ed edition (September 1, 1998).

31.

Van Duzer

. (2012). Random qualitative validation: A mixed-methods approach to survey validation. Retrieved from http://www.eric.gov

32.

Weaver

Yun

S. H.

(2010). Social work students’ attitude toward poverty in the new economy: A multivariate analysis. US-China Education Review, 7, 34–46.

33.

Weaver

Yun

S. H.

(2011). Analyzing the attitude of undergraduate students toward poverty and impoverished persons: Does social work education make a difference? Journal of Teaching in Social Work, 31, 3–20.

34.

Weiss

(2006). Factors associated with interest in working with the poor. Families in Society, 8, 385–394.

35.

Yang

(2010). Making sense of statistical methods in social research. Thousand Oaks, CA: Sage.

36.

Yun

S. H.

Weaver

(2010). Development and validation of a short form of the attitudes toward poverty scale. Advances in Social Work, 11, 174–187.

37.

Zygmunt-Fillwalk

(2009). Living the questions: Deconstructing interdisciplinary higher education and the subject of poverty through a “community of truth.” Journal of Poverty, 13, 214–232.