Content uploaded by Jason T Siegel
Author content
All content in this area was uploaded by Jason T Siegel on Feb 17, 2018
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=hjpa20
Journal of Personality Assessment
ISSN: 0022-3891 (Print) 1532-7752 (Online) Journal homepage: http://www.tandfonline.com/loi/hjpa20
Measuring Prosociality: The Development of a
Prosocial Behavioral Intentions Scale
Rachel Baumsteiger & Jason T. Siegel
To cite this article: Rachel Baumsteiger & Jason T. Siegel (2018): Measuring Prosociality: The
Development of a Prosocial Behavioral Intentions Scale, Journal of Personality Assessment, DOI:
10.1080/00223891.2017.1411918
To link to this article: https://doi.org/10.1080/00223891.2017.1411918
Published online: 15 Feb 2018.
Submit your article to this journal
View related articles
View Crossmark data
Measuring Prosociality: The Development of a Prosocial Behavioral Intentions Scale
Rachel Baumsteiger and Jason T. Siegel
School of Social Science, Policy, and Evaluation, Claremont Graduate University
ARTICLE HISTORY
Received 19 June 2017
Revised 19 October 2017
ABSTRACT
Prosociality is a critical issue in behavioral research. In this investigation, we developed a measure of
prosocial behavioral intentions. Qualitative responses from two surveys (nD465) and items from existing
measures were used to generate a list of prosocial behaviors in which people might intend to engage. We
factor analyzed responses to these items (nD319) and retained the most common and representative
items. The new measure demonstrated adequate internal consistency (nD247, 147; aD.81, .83);
convergent validity with past prosocial behavior (rD.51, .43), moral identity (rD.50, .55), and materialism
(rD–.30, –.20). The instrument also predicted prosocial behavior while controlling for a prior measure of
prosocial intentions, Exp(B) D1.99, Wald D10.59, pD.001, thereby demonstrating incremental predictive
validity. This 4-item scale could be used across contexts to advance the study of prosociality.
The beloved child television host, Mr. Rogers, is famously
quoted for relaying his mother’s advice: “[When bad things hap-
pen], look for the helpers. There will always be helpers.”Indeed,
there is a thriving literature on the ways in which people volun-
tarily and intentionally help other people. These actions are
referred to as prosocial behavior (Batson & Powell, 2003). Exam-
ples include volunteering at a food bank, donating money to
charity, registering to be an organ donor, and comforting friends
after they lose their job. Prosocial behavior is critical to the study
of individual differences, development, well-being, interpersonal
relationships, and group functioning (e.g., Batson & Powell,
2003; Pashak & Laughter, 2012; Pavey, Greitemeyer, & Sparks,
2011,2012; Szreter & Woolcock, 2004; Van Tongeren, Green,
Davis, Hook, & Hulsey, 2016). Therefore, it is imperative to
have reliable and accurate prosociality measures. This investiga-
tion aims to complement previous literature by creating and val-
idating a new measure of prosocial intentions.
Existing approaches
The two main approaches for assessing prosocial behavior are to
observe behavior directly and to measure behavioral intentions.
Direct measures of prosocial behavior have been used as the
main outcome of several social psychology experiments. For
instance, in the classic Good Samaritan study, participants were
asked to complete a task in another building, which required
them to walk past a staged confederate. The confederate was
slouched over, coughing, with his eyes closed. Participants’
responses were rated from 0 (least helpful)to5(most helpful)
based on the extent to which they noticed the person in need,
asked if the person wanted help, and insisted on helping (Darley
& Batson, 1973). In another study, participants heard what they
believed was a person in an adjacent room expressing physical
distress and asking for help. Prosociality was operationalized as
the amount of time that passed before participants left their seats
to help (Darley & Latan
e, 1968). Alternatively, some studies
assess prosociality by asking participants to divide earnings
between themselves and another participant. People who sacri-
fice some of their earnings to pay another person are considered
to be acting prosocially (e.g., Kahneman, Knetsch, & Thaler,
1986; Murphy & Ackermann, 2014).
Direct behavioral measures have significant merits (see
Crano, Brewer, & Lac, 2014). However, such behavioral mea-
surement is not always possible due to contextual limitations or
costs. As such, many researchers examine prosocial intentions.
Prosocial intentions reflect a person’s readiness to help others
(e.g., Agerstr€
om & Bj€
orklund, 2009). As the theory of planned
behavior describes, intentions are based on people’s attitudes,
perceived behavioral control, and subjective norms, and are a
direct antecedent of behavior (Ajzen, 1991). A meta-analysis of
47 experiments revealed that moderate-to-large changes in pro-
social intentions (dD.66) predict small-to-moderate changes in
health behavior (Webb & Sheeran, 2006), as well as intentions to
engage in charitable giving (Smith & McSweeney, 2007).
Unfortunately, even though measuring intentions can serve
as a useful proxy for assessing prosocial behavior, we were
unable to find any commonly used, validated measures of proso-
cial intentions. Rather, it appears scholars have measured proso-
cial intentions by creating new items for each study, or
modifying items from scales with similar purposes. For example,
Pavey et al. (2011) asked participants to indicate their intentions
to perform five actions such as “Give money to charity”or “Go
out of your way to help a stranger in need”over the next 6 weeks.
Each item was rated on a 7-point Likert scale from 1 (definitely
will not do this)to7(definitely will do this). In another study, the
same researchers (Pavey, Greitemeyer, & Sparks, 2012) modified
CONTACT Rachel Baumsteiger rachel.baumsteiger@cgu.edu Claremont Graduate University, Psychology, 150 E. 10th Street, Claremont, CA 91711.
© 2018 Taylor & Francis
JOURNAL OF PERSONALITY ASSESSMENT
https://doi.org/10.1080/00223891.2017.1411918
the self-report altruism scale (Rushton, Chrisjohn, & Fekken,
1981), which is a 20-item measure designed to assess previous
prosocial behavior. The modified scale asked participants to
indicate their intentions to enact six prosocial behaviors in the
next 2 weeks on a 5-point Likert scale, with responses ranging
from 1 (definitely will not)to5(definitely will). Items included
actions such as, “Do volunteer work for a charity”and “Give up
my time to do something for the community.”
Both of the scales just described performed admirably, and
both demonstrate that it is possible to assess prosocial inten-
tions with a small set of self-report items. However, we believed
that it would be beneficial to develop a new scale that builds on
the strengths of these measures and integrates knowledge
across the existing literature. In short, our goal was to create a
brief, validated, prosocial intentions scale that would be appro-
priate for research with adults. Developing this measure could
help to standardize findings throughout the field, which in turn
could contribute to a more comprehensive understanding of
prosocial behavior.
Current studies
This investigation drew on the strengths of previous work (e.g.,
Pavey et al., 2011,2012; Penner, 2002; Rushton et al., 1981)to
develop and validate a brief measure of prosocial behavioral
intentions for adults. The overall approach to the current scale
validation was based on recommendations by Clark and Wat-
son (1995): We began with a large, inclusive list of potential
items, narrowed that list down based on a single dimension,
and then tested the relationships between the new scale and
scales of similar constructs to evaluate how accurately the scale
measures the target construct.
In the first study, qualitative data and items from existing
measures were evaluated and combined to generate a list of com-
mon prosocial behaviors. In Study 2, participants indicated how
likely they would be to perform each of the prosocial behaviors
identified in Study 1. Their responses were factor analyzed to
determine which items were most representative. These items
were then given to participants along with conceptually related
scales to evaluate psychometric characteristics of a new prosocial
intentions measure. Finally, the same survey was administered
to another sample along with items that were used to assess pro-
social intentions in a previous study (Pavey et al., 2011) to assess
whether the Prosocial Behavioral Intentions Scale (PBIS) pro-
vided incremental predictive validity. The overarching research
questions of this investigation were as follows:
1. What forms of prosocial behavior are most common,
generalizable, and representative? (Studies 1 and 2)
2. Does the new measure of prosocial intentions exhibit
adequate internal consistency, convergent validity, and
predictive validity? (Studies 3 and 4)
3. Does the new measure of prosocial intentions possess
incremental predictive validity in relation to a prior
assessment strategy? (Study 4)
Study 1
The goal of this study was to generate a list of common proso-
cial behaviors to create a new prosocial intentions scale.
Examples were taken from laypeople’s spontaneous responses
to how they could help or have helped other people (Study 1a),
as well as from existing measures that assess similar constructs
(Study1b). This inductive-deductive approach was used to help
generate a large, inclusive list of behaviors that are both aligned
with the current scientific conceptions of prosociality and are
grounded in real-world experiences.
Method
Study 1a
Participants.The sample for the qualitative analyses included
465 U.S. adults recruited from Amazon’s Mechanical Turk
(MTurk). Research indicates that MTurk workers are more
diverse than most undergraduate student samples and tend to
provide reliable data (Azzam & Jacobson, 2013; Casler, Bickel, &
Hackett, 2013;Goodman,Cryder,&Cheema,2013). Participants
were between the ages of 18 and 67 (MD28.83, SD D9.39).
Approximately half were female (51%) and half were male
(49%). Most participants were White (59%), a large number were
Asian (22%), and the remaining people were African American
(10%), Hispanic or Latino (6%), or other ethnicities (3%).
Materials.Participants were asked to respond to one of
three prompts: (a) Describe how you have helped others in the
past week, (b) Describe how you have helped others in the past,
or (c) Describe how you could help someone in the future.
Procedure.Two separate surveys were administered. As a
part of a larger project on the antecedents of prosocial behavior,
two survey links were posted on MTurk.com. People who fol-
lowed these links were led to surveys on Qualtrics.com. After
providing consent, participants were asked to respond to
prompt a or b (survey 1) or a or c (survey 2). In total, 217 par-
ticipants responded to prompt a, 216 responded to prompt b,
and 122 responded prompt c. Next, they answered additional
questions about prosocial behavior and received compensation.
We combined data from both surveys for analyses for the cur-
rent study. Responses to the writing prompts were compiled
for review.
Study 1b
Procedure. A search was conducted in PsycINFO for any scale
containing the key words, prosocial, helping, or altruistic in com-
bination with the terms scale, measure, battery, and question-
naire. This search led to the identification of several measures
that assess prosocial attitudes. For example, the altruistic attitudes
scale contains items such as, “I enjoy doing things for others”
(Kahana, Bhatta, Lovegreen, Kahana, & Midlarsky, 2013). Other
scales included specificexamplesofprosocialbehaviorssuchas
“I have given directions to a stranger”and “I have done volunteer
work for a charity”(Penner, 2002; Rushton et al., 1981), which
were more aligned with our research goals.
Results
First, we combined items from existing measures with examples
of prosocial behavior that people described in qualitative
responses. Next, we simplified items to their sentence stems
and standardized verb tenses. For example, “I have given direc-
tions to a stranger”was simplified to “Give directions to a
2 BAUMSTEIGER AND SIEGEL
stranger.”We then merged similar behaviors such as “Help my
friend with a homework assignment”and “Help my sibling do
their homework.”We also removed items that described
actions that some people are not physically capable of doing
such as shoveling snow and donating blood. These steps
resulted in approximately 50 items.
Given that the goal was to find items that were as representa-
tive as possible, and because answering so many similar items
could be taxing on participants, we attempted to narrow down
this list further. Items were categorized according to several
themes that emerged from the data: the magnitude of help (small,
medium, large), the recipient of help (a close friend or family
member, an acquaintance, a stranger, or the community at large),
whether or not the effects of helping could be observed directly,
and the cost of helping (time, effort, resources). Items that were
similar on all dimensions were combined and represented by a
more general description. For instance, “Letting someone borrow
my car,”“Lending someone my lawn mower,”and “Letting
someone use my phone”each involves allowing someone to use
something of value. Therefore, those three items were combined
intothemoregeneralitem,“Lend someone an item that I care
about, like a car or a favorite jacket.”In addition to reducing the
number of items, this strategy led us to create more general items,
which might reduce variance in scores due to factors that are
unrelated to prosociality. For instance, in the previous example,
some people might have a car but not a lawn mower, and vice
versa, so the specificitemusedcouldinfluence which people
score higher on prosociality overall. In contrast, responding to
the general behavior of lending resources might better reflect the
underlying prosocial behavioral intention. Thus, we believed that
making items more general might enhance how well they repre-
sent overall prosocial intentions for people across different types
of contexts. The final list included 20 prosocial behaviors.
Discussion of Study 1
The goal of this study was to produce a list of prosocial
behaviors that are common and accessible to all people. To
identify these behaviors, we evaluated actions that were pro-
posed by both laypeople (through qualitative responses) and
researchers (through existing measures). There was high
consensus within and between these sources in terms of the
kinds of things people do to help others. To limit items to
generalizable behaviors, we evaluated each item in terms of
whether it was accessible to adults of different ages, regions,
social class, physical and psychological capabilities, ideologi-
cal stances, and personal resources. These steps led to a list
of 20 statements (see Table 1). Given that including all
items would lead to a relatively lengthy measure, the next
step was to reduce this list by determining which items
were most representative of the overarching construct of
prosocial behavioral intentions.
Study 2
The purpose of Study 2 was to evaluate which prosocial behav-
iors identified in Study 1 were most representative. More spe-
cifically, we sought to narrow down the initial 20 items to the
smallest number of items possible that accounted for most of
the variance in scores.
Method
Participants.Thesampleconsistedof319MTurkworkers
from the United States. This sample size satisfies the recommen-
dation that there should be at least 10 participants for every
potential item tested (20) in an exploratory factor analysis
(Garson, 2008). Participants’ages ranged from 19 to 77
Table 1. Study 2 factor loadings.
S1 S2 Communalities
Item F1 F2 F3 F4 S1 S2
1. Comfort someone I know after they experience a hardship .75 .86 .56 .73
2. Care for someone’s child, animal, or home for free .74 .54 .60
3. Offer to help someone I know with a difficult project .73 .60 .53 .58
4. Mentor a younger or less experienced person .71 .51 .51
5. Do a task/chore for my friend, family member, colleague, etc. .71 .59 .51 .56
6. Help care for a sick friend or relative .70 .78 .49 .61
7. Volunteer for a fundraiser .69 .82 .47 .73
8. Help out in a school/religious/community organization .69 .63 .47 .65
9. Volunteer for a philanthropic organization .66 .87 .44 .75
10. Donate old books, clothes, toys, etc. to charity .66 .51 .43 .47
11. Donate my change to a charity .65 .42 .47
12. Assist a stranger with a small task (e.g. help them carry groceries, watch their things
while they use the restroom)
.64 .73 .41 .64
13. Help someone I know move their belongings to a new residence .63 .52 .40 .51
14. Make efforts to help a stranger who endured a hardship (send money or a kind message) .62 .38 .57
15. Send someone I know a supportive message .62 .82 .38 .55
16. Help a stranger find something they lost, like a piece of jewelry or a pet .60 .86 .37 .70
17. Give directions to a stranger .59 .77 .34 .61
18. Lend someone an item that I care about, like a car or a favorite jacket .52 .57 .27 .56
19. Participate in a social/political movement .52 .73 .27 .55
20. Recycle -.61 .19 .59
Eigenvalues (unrotated) 8.38 8.38 1.38 1.11 1.05
Total variance explained 41.91% 59.63%
Note. S1 DSolution 1, S2 DSolution 2, F1 DFactor 1, F2 DFactor 2, F3 DFactor 3, F4 DFactor 4. Values <.50 are omitted. As explained shortly, items shown in italics
were originally retained, but removed from the scale post-hoc. Items shown in bold were retained for the final scale.
PROSOCIAL BEHAVIORAL INTENTIONS SCALE 3
(MD35.92, SD D12.06). There were more females (65%) than
males (34%), and there were two nonbinary gendered partici-
pants (1%). Most participants were White (68%), and the
remaining participants were African American (10%), Asian
(9%), Hispanic or Latino (7%), or other ethnicities (5%).
Materials
Prosocial intentions. Participants responded to the 20 proso-
cial behavioral intention items identified in Study 1. Instruc-
tions for intentions scales sometimes ask respondents to report
how likely they would be to perform behaviors within a given
time frame, such as the next week (e.g., Pavey et al., 2011).
However, limiting the time frame could lead people to respond
based on their perceptions of the likelihood of each situation
presenting itself rather than their willingness to perform that
behavior. Therefore, we worded the instructions to ask
respondents to assume that the opportunity to enact each
behavior presents itself and set the time frame to the general
future. Each item was rated on a 7-point Likert scale ranging
from 1 (definitely would not do this)to7(definitely would do
this). Items included behaviors such as, “Care for a sick friend
or relative.”All items were coded positively, with higher scores
indicating stronger intentions to perform the behaviors.
Attention check. An additional item was embedded within
the prosociality scale to evaluate whether participants were pay-
ing attention. This item instructed participants to “Please select
2 for this item.”This procedure follows recommendations for
ensuring data quality for online surveys (Berinsky, Margolis, &
Sances, 2014).
Procedure
Participants were recruited from MTurk.com and completed
the survey posted on Qualtrics.com. The survey took approxi-
mately 4 min to complete. The data were factor analyzed to
assess which items accounted for most variance in scores. This
information was used to evaluate which items would be
retained for the final scale.
Results
Preliminary analyses
The original database included 331 cases. Before conducting
analyses, we removed 12 responses from people who answered
the attention check incorrectly. No cases were removed for
missing data. Next, we checked the normality of data. All distri-
butions were negatively skewed (skewness ranged from –2.05 to
–0.33). No outliers were apparent. In addition, there were no
nonlinear relationships among items. Item correlations ranged
from .10 to .67, with all correlations significant at p<.001. Fur-
ther, the data satisfied additional statistical assumptions for an
exploratory factor analysis (Kaiser-Meyer-Oklin D.92, Bar-
tlett’s test p<.001).
Main analyses
We performed a principal components factor analysis to deter-
mine which items captured the most variance in overall proso-
cial behavioral intention scores. Based on the scree plot and on
the assumption that items represent the single factor of
prosocial behavioral intentions, we evaluated a one-factor solu-
tion (Solution 1; Eigenvalue D8.38) that accounted for 41.91%
of the total variance in scores. We also explored an alternative
factor solution (Solution 2) based on the suggestion of consid-
ering factors with Eigenvalues greater than one (Kaiser, 1960).
We performed an oblique rotation to improve the interpretabil-
ity of factors on Solution 2 because we expected to find overlap
among factors. This led to a four-factor solution (Eigenvalues
D8.38, 1.38, 1.11, and 1.05) that accounted for 59.63% of the
total variance cumulatively. Correlations among the four fac-
tors ranged from rD.06 to rD.54. Table 1 displays pattern
matrix factor loadings and communalities for both solutions.
After reviewing which items loaded on each of the four factors
in Solution 2, we labeled the factors (a) helping close others, (b)
donating time or resources, (c) helping a stranger, and (d)
items that did not fit in as well with the others.
Item selection
Items were evaluated for inclusion in the final scale based on
several statistical and theoretical considerations. First, we
removed items that were not strongly correlated with the other
items because they presumably did not measure the underlying
construct as well as the others (i.e., 18 and 20). Second, given
that the first three factors in Solution 2 seemed to present a
meaningful distinction (the recipient of help), we ranked the
items that scored highest on each factor. Next, we evaluated the
factor loadings and communalities of each item to further rank
which items within each factor accounted for the most overall
variance. Finally, we considered the additional dimensions
identified in Study 1: magnitude of help, costs involved in help-
ing, and whether the effects of help could be observed directly.
We selected the items that differed on these dimensions so the
resulting scale would assess people’s intentions to engage in
diverse types of prosocial behavior. An initial evaluation of the
resulting scale indicates that it has adequate internal consis-
tency (aD.76) with composite scores ranging from 2.00 to
7.00 (MD5.90, SD D1.01, skewness D–1.06, kurtosis D1.00).
Discussion of Study 2
In this study, we used information from analyses to choose
items that differed based on features such as the recipient of
help, magnitude of help, and cost of helping. We then selected
the two items that that captured the most variance within the
“helping close others”factor and the two items that captured
the most variance within the “helping strangers”factor
1
(Eisenberg & Spinrad, 2014). This new measure is referred to
from here on out as the PBIS. The next step in developing this
measure was to determine whether it accurately and reliably
assesses prosocial behavioral intentions.
1
We originally retained eight items for the new scale, which was then used in sub-
sequent studies. After further consideration, we realized that we overlooked the
fact that some people do not have the resources to donate their time or used
items. Based on this conclusion, we modified the scale to exclude the two items
that described volunteering time and the one item about donating used items
(e.g., toys, clothes, books). We also removed an additional item to maintain the
balance between items that referred to helping close others and strangers. The
four items that were removed are italicized in Table 1. From here forward, we
report findings based on the modified four-item scale. Results using the eight-
item scale are available from the authors.
4 BAUMSTEIGER AND SIEGEL
Study 3
The aim of Study 3 was to evaluate psychometric characteristics
of the PBIS. First, we expected the scale to exhibit adequate
internal consistency (H1). Next, we tested correlations between
the PBIS and related constructs. Given that previous work links
prosocial behavior in the past, present, and future (Ouellette &
Wood, 1998), we hypothesized that the PBIS would be posi-
tively correlated with past prosocial behavior (H2). Similarly,
individuals who possess a strong moral identity—people who
consider doing the right thing as central to their sense of self
(Aquino & Reed, 2002; Frimer & Walker, 2009)—are more
likely than others to exhibit prosocial behaviors (Sage, Kavus-
sanu, & Duda, 2006; Winterich, Aquino, Vikas, & Swartz,
2013). Thus, we hypothesized that the PBIS would be positively
correlated with moral identity (H3). Furthermore, prosocial
behavior involves a concern for others, which is negatively
related to materialism, or a desire to possess expensive things
to boost one’s social status (Freund & Blanchard-Fields, 2014).
Therefore, we hypothesized that the PBIS would be negatively
correlated with materialism (H4). Finally, based on the link
between prosocial intentions and behavior (Ajzen, 1991; Smith
& McSweeney, 2007), we hypothesized that PBIS scores would
predict prosocial behavior (H5).
Method
Participants
The final sample included 247 MTurk workers. A power analy-
sis using G
Power (Faul, Erdfelder, Lang, & Buchner, 2007)
indicates that the sample size provides sufficient power to
detect a large or moderate effect size in correlational analyses.
Participants were between the ages of 18 and 71 (MD35.60,
SD D11.46), with approximately half (51%) of the sample con-
sisting of females and half (49%) consisting of males. Most par-
ticipants were White (81%) and the remaining participants
were African American (6%), Asian (6%), Hispanic or Latino
(5%), mixed (2%) or other ethnicities (<1%).
Materials
Prosocial behavioral intentions. One item was changed to
make it more universal. Specifically, the text in the item starting
with “Help a strange find something they lost”was changed
from “like a piece of jewelry or a pet”to “like their key or a
pet.”The PBIS created in Study 2 was used to assess prosocial
behavioral intentions. See earlier for a description and the
Appendix for the full scale.
Past prosocial behavior. The self-reported altruism sub-
scale of the prosocial personality battery (Penner, 2002) was
used to measure the frequency of previous prosocial behavior.
Respondents are asked to report how often they have carried
out five helping behaviors in the past. Items include actions
such as, “I have helped carry a stranger’s belongings”and “I
have let a neighbor whom I didn’t know too well borrow an
item of some value (e.g., tools, a dish, etc.).”This subscale
exhibits adequate internal consistency (aD.73; Penner, 2002).
Moral identity. The Self-Importance of Moral Identity
Scale (Aquino & Reed, 2002) is a 10-item self-report measure
of how much a person bases his or her self-concept on moral
values, beliefs, and behavior. The instructions ask respondents
to think about a person who possesses moral qualities (e.g.,
honest, kind), and then rate their disagreement or agreement
with statements such as, “I strongly desire to have these charac-
teristics”and “Being someone who has these characteristics is
an important part of who I am.”Each item is rated on a 7-point
Likert scale, with responses ranging from 1 (strongly disagree)
to 7 (strongly agree). Higher scores on this scale indicate a
stronger moral identity. Prior studies indicate that this scale
has demonstrated adequate internal consistency (aD.73–.82;
Aquino & Reed, 2002).
Materialism. The Materialism Scale—Modified (Sirgy et
al., 2012) is a nine-item self-report measure of the importance of
owning expensive luxury items to a person’s general satisfaction.
Respondents are asked to rate their disagreement or agreement
with each item, which includes statements such as, “Having lux-
ury items is important to a happy life”and “I love to buy new
products that affect status and prestige.”Each item is rated on a
7-point scale from 1 (strongly disagree)to7(strongly agree). Pre-
vious work suggests that this scale possesses adequate internal
reliability (aD.93; Sirgy et al., 2012).
Prosocial behavior. Respondents were presented with two
additional items at the end of the survey. The instructions stated
that the next two questions were optional and that the partici-
pant could skip them with no penalty. It was emphasized that
taking the extra time to fill these out would help the researchers
with their future work. The word optional was presented in capi-
tal letters and highlighted in yellow to emphasize that these
questions were not required. The two questions were, “How
would you define a good life?”and “How would you define mor-
ality?”Under each question, there was a text box for participants
to type their responses. This behavior aligns with the types of
actions assessed in the PBIS, especially the item that asks partici-
pants to rate their intentions to help a stranger with a small task.
Researchers have used similar strategies to assess prosocial
behavior in a survey (e.g., Siegel, Thomson, & Navarro, 2014;
Thomson, Nakamura, Siegel, & Csikszentmihalyi, 2014).
Attention check. An additional item was embedded within
the Materialism Scale to assess whether participants were pay-
ing attention. This item instructed participants to “Please select
1 for this item.”
Procedure
A link to the survey was posted on MTurk.com and the survey
was on Qualtrics.com. A restriction was set so people from Study
2 could not enter this study. People who clicked on
the link and consented to participate went on to complete the
measures in random order, and then were presented with the
request to answer additional questions. The survey took approxi-
mately 5.5 min to complete. We created composite scores for
each measure. For all studies, when computing scale composites,
incomplete data was addressed using mean imputations. Next, we
calculated one-tail correlations to evaluate relationships among
constructs. We also used a ttest to compare scores on the PBIS
between people who behaved prosocially and those who did not.
In response to a reviewer’s recommendation, we conducted a
word count on the open-ended responses that represented proso-
cial behavior and computed correlations between the PBIS and
number of words people wrote for the voluntary questions.
PROSOCIAL BEHAVIORAL INTENTIONS SCALE 5
Results
Preliminary analyses
The original database included 250 cases. Data screening
revealed that three participants failed the attention check, so
their data were deleted prior to analyses. There were no other
suspicious cases. The distributions of the main variables were
all normally distributed, with skewness ranging from –1.26 to
0.60 and kurtosis ranging from –0.70 to 2.13. Scores were nor-
mally distributed on the PBIS (MD5.73, SD D1.10), moral
identity scale (MD5.05, SD D.88), measure of past prosocial
behavior (MD4.91, SD D1.29), and Materialism Scale—Modi-
fied (MD2.65, SD D1.41). Approximately half of participants
(56%) completed the additional survey questions.
Main analyses
The PBIS demonstrated adequate internal consistency, aD.81. It
was positively correlated with past prosocial behavior, rD.51,
p<.001, and moral identity, rD.50, p<.001. In contrast, it was
negatively correlated with materialism, rD–.30, p<.001. See
Table 2 for a correlation matrix of the main variables. The PBIS
was also related to prosocial behavior, such that those who
behaved prosocially by responding to the optional questions
scored significantly higher on the PBIS (MD5.94, SD D.96)
than those who did not respond (MD5.46, SD D1.21), t(200) D
3.36, p<.001, dD.44. Furthermore, there was a small, significant
correlation between the PBIS and the number of words people
wrote for the voluntary questions, rD.24, p<.001.
Discussion of Study 3
All hypotheses were supported: The PBIS exhibited adequate
internal consistency (H1), was significantly positively corre-
lated with past prosocial behavior (H2) and moral identity
(H3), and was significantly negatively correlated with material-
ism (H4). Most important, participants who scored higher on
the PBIS were significantly more likely than those who scored
lower to answer additional survey questions for no payment
(H5). They also provided longer responses for those questions.
Altogether, the results of this study provide evidence that the
PBIS has good psychometric properties, and thus it is ade-
quately assessing prosocial intentions.
As noted earlier, this research endeavor sought to bring
together and build on prior investigations that used measures
of prosocial intentions. One example of prior scholarship that
aided us in this regard was an instrument created by Pavey et
al. (2011). Although these items were not formally named in
their original publication, we refer to them as the Prosocial
Intentions Measure (PIM). As a final validity assessment, we
assessed the incremental predictive validity of the PBIS in rela-
tion to the PIM.
Study 4
The purpose of this study was to evaluate the incremental pre-
dictive validity of the PBIS in relation to the PIM (H1). A sec-
ondary goal was to assess the consistency in correlations
between the PBIS and measures of related constructs. Accord-
ingly, we hypothesized that the PBIS would be positively related
to past prosocial behavior (H2) and moral identity (H3), and
negatively correlated to materialism (H4). We also conducted
an auxiliary analysis comparing the correlations of the PBIS to
those of the PIM in regard to past prosocial behavior, moral
identity, and materialism.
Method
Participants
The final sample included 147 MTurk workers. A power
analysis using G
Power (Faul et al., 2007)suggestedthat
this sample size provided sufficient power to detect effect
sizes comparable to those found in Study 3. Participants’
ages ranged from 18 to 76 (MD35.40, SD D11.74). There
were slightly more females (59%) than males (41%) and
there was one transgendered person (<1%). Most partici-
pants were White (74%). Others were African Americans
(10%), Hispanic (8%), Asian (7%), Native American (1%),
or PacificIslander(1%).
Materials
The survey from Study 3 included the PBIS (developed through
Studies 1 and 2), past prosocial behavior scale (Penner, 2002),
moral identity scale (Aquino & Reed, 2002), Materialism Scale–
Modified (Sirgy et al., 2012), a request to answer two additional
open-ended questions for no payment (which served as a mea-
sure of prosocial behavior), and demographic questions such as
age, gender, and ethnicity. In addition, the survey included the
PIM (Pavey et al., 2011). As described previously, the PIM asks
respondents to indicate how likely they are to perform five
behaviors in the next 6 weeks.
Procedures
The procedure was identical to Study 3, with the exception
that the PIM was added to the survey. The survey took
approximately 8 min to complete. Responses were averaged
across each measure to create composite scores. We also
counted the number of words people wrote in response to the
voluntary questions. A binary logistic regression was used to
evaluate whether the PBIS predicted prosocial behavior more
strongly than the PIM. One-tailed correlations were then
computed between the main variables to examine the relation-
ships between the PBIS and measures of related constructs. As
an auxiliary analysis, we compared correlations between the
PBIS and PIM with the measures used to assess convergent
validity.
Table 2. Study 3 correlation matrix.
1. 2. 3. 4.
1. PBIS .81
2. Past prosocial behavior .51
.75
3. Moral identity .50
.36
.78
4. Materialism ¡.30
¡.09 ¡.02 .96
Note. PBIS DProsocial Behavioral Intentions Scale. Reliabilities are listed in the
diagonal.
p<.01.
p<001.
6 BAUMSTEIGER AND SIEGEL
Results
Preliminary analyses
Of the original 150 respondents, three cases were removed
prior to analyses because those people failed the attention
check. All other data were retained. There were no outliers on
the main variables. Scores were normally distributed on the
PBIS (MD5.86, SD D1.10), PIM (MD4.87, SD D1.39),
moral identity scale (MD4.89, SD D1.09), measure of past
prosocial behavior (MD4.22, SD D1.54), and materialism
scale (MD2.66, SD D1.63), with skewness ranging from –1.25
to 0.90 and kurtosis ranging from –0.69 to 2.29. A little over
half of participants (59%) completed the additional survey
questions, which served as the measure of prosocial behavior.
Main analysis
A binary logistic regression was conducted to test whether the
PBIS predicted prosocial behavior beyond what was predicted
by the PIM. Prosocial behavior was the dependent variable (0
Ddid not help,1Ddid help). The PIM was entered in Step 1.
By itself, the PIM was not a statistically significant predictor of
prosocial behavior, Exp(B) D1.27, Wald D3.66, pD.06. The
PBIS was then entered in Step 2. At this step, the PIM was not
a statistically significant predictor of prosocial behavior, Exp(B)
D.95, Wald D.10, pD.76, but the PBIS was, Exp(B) D1.99,
Wald D10.59, pD.001. Based on the odds ratios, people who
scored 1 point higher on the PBIS had approximately twice the
odds, on average, of behaving prosocially when taking the PIM
into account. To investigate further, we conducted a second
binary logistic regression with the PBIS entered at Step 1 and
the PIM entered at Step 2. At Step 1, the PBIS accounted for a
significant portion of variance in prosocial behavior, Exp(B) D
1.92, Wald D13.17, p<.001. Similarly, the PBIS accounted for
a significant amount of variance in prosocial behavior at Step 2,
Exp(B) 1.99, Wald D10.59, pD.001, whereas the PIM did not,
Exp(B) D.95, Wald D.10, pD.76. Taken together, these find-
ings indicate that the PBIS provides incremental predictive
validity above and beyond the PIM.
The PBIS was internally consistent (aD.83) and was posi-
tively correlated with past prosocial behavior, rD.43, p<.01,
and moral identity, rD.55, p<.01, and negatively correlated
with materialism, rD–.20, p<.01. The PBIS was also related
to prosocial behavior such that participants who helped the
researcher scored significantly higher on the PBIS (MD6.16,
SD D.82) than those who did not help (MD5.44, SD D1.31),
t(90) D3.74, p<.001, dD.66. In contrast, the relationship
between the PBIS and the length of qualitative responses was
significant, rD.16, pD.02.
Auxiliary analysis
As an additional analysis, we used a Fisher’srto ztransfor-
mation to compare correlations between the PBIS and the
PIM with regard to past prosocial behavior, moral identity,
and materialism. Correlations between the PBIS and PIM
did not differ significantly for moral identity, zD.90, pD
.18. The correlation between the PBIS and past prosocial
behavior was significantly smaller than the correlation
between the PIM and past prosocial behavior, zD2.25, p
D.01. The correlation between the PBIS and materialism
was significantly larger than the correlation between the
PIM and materialism, zD1.97, pD.02. See Table 3 for a
correlation matrix.
Discussion of Study 4
The main goal of this study was to test the incremental pre-
dictive validity of the PBIS in comparison to the PIM. The
first hypothesis was supported; The PBIS predicted proso-
cial behavior above and beyond what the PIM predicted.
This could be because the PBIS has a broader time frame
than the PIM, which allows it to capture prosociality rather
than expectations of future helping opportunities. On the
other hand, it could be because the PBIS avoids potential
problems associated with assessing prosociality based on
intentions to contribute one’s resources, which all people
might not be equally able to do.
The second goal of this study was to assess whether the PBIS
exhibited similar psychometrics properties in a new sample.
Comparisons reveal similarities in the distributions of PBIS
scores across studies (Study 3 MD5.73, SD D1.10; Study 4
MD5.86, SD D1.10). Likewise, the PBIS exhibited similar
internal consistency (Study 3 aD.81; Study 4 aD.83) and was
almost equally related to past prosocial behavior (Study 3 rD
.51; Study 4 rD.43), moral identity (Study 3 rD.50; Study 4
rD.55) and materialism (Study 3 rD–.30; Study 4 rD–.20)
in terms of both effect sizes and significance levels. These find-
ings suggest that the PBIS has high psychometrically stability.
General discussion
The purpose of this investigation was to develop and validate
a prosocial intentions scale that unites previous work and
could be used across research contexts. Data from the current
studies indicate that the PBIS is internally consistent; has
good convergent validity in terms of its relation to past proso-
cial behavior, moral identity, and materialism; and, predicts
prosocial behavior above and beyond a previous assessment
tool. The sizes and directions of relationships among these
constructs are consistent with previous work (Aquino &
Reed, 2002; Freund & Blanchard-Fields, 2014; Frimer &
Walker, 2009;Sageetal.,2006;Winterichetal.,2013). Fur-
thermore, the effect sizes for correlations between prosocial
intentions and behavior were small to medium (Study 3 dD
.44, Study 4 dD.66). These effect sizes align with, or exceed,
effect sizes found between other intentions and behaviors,
such as exercise (dD.39; Jones, Sinclair, & Courneva, 2003),
Table 3. Study 4 correlation matrix.
1. 2. 3. 4. 5.
1. PBIS .83
2. PIM (Pavey et al., 2011) .57
.85
3. Past prosocial behavior .43
.62
.84
4. Moral identity .55
.62
.43
.84
5. Materialism ¡.20
.03 .11 ¡.01 .97
Note. PBIS DProsocial Behavioral Intentions Scale; PIM Dprosocial intentions
measure. Reliabilities are listed in the diagonal.
p<.01.
p<001.
PROSOCIAL BEHAVIORAL INTENTIONS SCALE 7
sun protection (dD.43; Lescano, 1999), smoking (dD.12;
D’Onofio, Moskowitz, & Braverman, 2002), and donating
money to environmental organizations (dD.60; Hine & Gif-
ford, 1991; see Webb & Sheeran, 2006, for a meta-analysis).
Most important, the PBIS demonstrated incremental predic-
tive validity in comparison to the PIM (Pavey et al., 2011).
Taken together, these findings indicate the PBIS is a relatively
valid measure for assessing prosocial intentions.
In addition to demonstrating good psychometric properties,
the PBIS possesses several practical qualities that make it
appealing across various research contexts: It is easy to admin-
ister via pen and paper or via online surveys; is brief enough to
be included in surveys without adding too much length; con-
tains simple language that most people can comprehend; and
includes examples of behavior that are available to most people.
This measure could serve many purposes in research. For
example, it could be used to assess interindividual differences
in prosociality, to compute correlations between prosociality
and other constructs, to detect developmental differences in
prosociality, or to test the efficacy of interventions in promot-
ing prosocial intentions.
Despite its strengths, the PBIS is not appropriate for all
contexts. For instance, a person might rate themselves as less
likely to “Help care for a sick friend or relative”if they live in
isolation or do not have any living friends or relatives. Addi-
tionally, children and adolescents below age 14 likely exhibit
different types of prosocial behavior than those included in
the PBIS. It likely would be more useful to assess younger
people’s prosocial intentions by asking about actions such as
helping a classmate study or joining a student group that
aims to improve their school (see Scales & Benson, 2003). In
sum, the PBIS is not appropriate for every individual. None-
theless, it could be used in a broad range of research studies
involving typical adults.
The results of this research should be interpreted within the
context of its limitations. First, all four studies used responses
from MTurk workers. Although this population is shown to be
demographically diverse and to provide good data (Azzam &
Jacobson, 2013; Casler, Bickel, & Hackett, 2013; Goodman et
al., 2013), it is possible that their prosocial behavioral intentions
differ from other groups of people. For instance, older adults
are less likely than younger and middle-age adults to use tech-
nology such as online survey Web sites (Pew Research Center,
2017), and they might also be likely to envision themselves per-
forming different types of prosocial behaviors than their youn-
ger counterparts. Second, responses to the PBIS could be
influenced by people’s desire to appear prosocial—either to
themselves or to the researchers. Moving forward, it would be
useful to administer a measure of social desirability (see
Crowne & Marlowe, 1960) to gauge these effects. Third, some
of the items ask participants about two different behaviors (e.g.,
help care for a sick friend or a relative). Even though there are
advantages to this approach (e.g., some respondents might not
have a relative they would care for, others might not have a
friend they would care for), these items can also be considered
double-barreled (Dillman, 2000). To allow for the benefit
of this approach, but avoid potential harms, we recommend
adding a line to the instructions that reads, “If you are more
likely to complete one task (e.g., help a stranger find a key)
than another (e.g., help a stranger find a missing pet), please
respond to the task that you would be more likely to perform.”
Fourth, prosocial behavior was assessed with one specific act.
This behavior required very little time and effort, especially
given that the sample consisted of people who complete surveys
daily and the request was to answer two brief questions. Effects
might have been smaller or larger if prosocial behavior was
assessed with an action that involved more time, effort, and
consideration, which could reveal a clearer divide in rates of
prosociality. Finally, much of this scale development involved
subjective judgments. For example, there was no objective crite-
rion for selecting the number of items to retain in Study 2. We
made this decision by balancing considerations for adequate
representativeness and parsimony, but other researchers might
have selected more or fewer items.
This research illuminates several promising directions for
future investigation. One is to conduct further validation
testing on the current scale. For instance, researchers could
test whether the PBIS predicts other types of prosocial
behaviors, such as caregiving or registering to be an organ
donor (e.g., Siegel et al., 2010). Next, it would be useful to
test the PBIS among different samples. Given that MTurk
workers tend to be slightly more liberal and educated than
the general U.S. population (Goodman et al., 2013), it would
be useful to test the PBIS in non-MTurk samples. Further-
more, norms related to prosociality—such as expectations
for children to care for aging parents—often vary across cul-
tures (e.g., Lee & Sung, 1997). Thus, it would be valuable to
test the psychometric qualities of the PBIS across English-
speaking countries outside of the United States. Another
direction for future research is to administer the PBIS to the
same people across multiple time points to determine
whether it captures intraindividual differences that emerge
across time or as the result of an intervention. Because pro-
social intentions are often measured as a proxy for behavior,
it would also be useful to see how well the PBIS predicts
additional types of helping behavior such as volunteering on
a long-term basis or comforting a friend. Finally, although
the brevity and simplicity of this scale make it useful for
many studies, it would also be valuable to develop a more
extensive measure that obtains information across key
dimensions of prosocial behavior, such as the beneficiary of
help. A multidimensional scale could be particularly useful
for studies in which prosocial intentions are the main con-
struct. Taken together, these studies provide evidence for a
valid and practically useful measure of prosocial behavioral
intentions. Continuing this line of research can enhance the
accuracy of prosociality measurement, which can contribute
to a better understanding of prosocial behavior.
References
Agerstr€
om, J., & Bj€
orklund, F. (2009). Temporal distance and moral con-
cerns: Future morally questionable behavior is perceived as more
wrong and evokes stronger prosocial intentions. Basic and Applied
Social Psychology, 31, 49–59. doi:10.1080/01973530802659885
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior
and Human Decision Processes, 50, 179–211. doi:10.1016/0749-5978
(91)90020-T
8 BAUMSTEIGER AND SIEGEL
Aquino, K., & Reed, A. (2002). The self-importance of moral identity. Jour-
nal of Personality and Social Psychology, 83, 1423–1440. doi:10.1037/
0022-3514.83.6.1423
Azzam, T., & Jacobson, M. R. (2013). Finding a comparison group: Is
online crowdsourcing a viable option? American Journal of Evaluation,
34, 372–384. doi:10.1177/1098214013490223
Batson, C. D., & Powell, A. A. (2003). Altruism and prosocial behavior. In
T. Millon & M. J. Lerner (Eds.), Handbook of psychology: Vol. 5: Person-
ality and social psychology (pp. 463–484). Hoboken, NJ: Wiley.
Berinsky, A. J., Margolis, M. F., & Sances, M. W. (2014). Separating the
shirkers from the workers? Making sure respondents pay attention on
self-administered surveys. American Journal of Political Science, 58,
739–753. doi:10.1111/ajps.12081
Casler, K., Bickel, L., & Hackett, E. (2013). Separate but equal? A compari-
son of participants and data gathered via Amazon’s MTurk, social
media, and face-to-face behavioral testing. Computers in Human
Behavior, 29, 2156–2160. doi:10.1016/j.chb.2013.05.009
Clark, L. A., & Watson, D. (1995). Constructing validity: Basic issues in
objective scale development. Psychological Assessment, 7, 309–319.
doi:10.1037/1040-3590.7.3.309
Crano, W. D., Brewer, M. B. & Lac, A. (2014). Principles and Methods of
Social Research. London, UK: Taylor-Routledge.
Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability
independent of psychopathology. Journal of Consulting Psychology, 24,
349–354. doi:10.1037/h0047358
D’Onofrio, C. N., Moskowitz, J. M., & Braverman, M. T. (2002). Curtailing
tobacco use among youth: Evaluation of Project 4-Health. Health Edu-
cation and Behavior, 29, 656–682. doi:10.1177/109019802237937
Darley, J. M., & Batson, C. D. (1973). From Jerusalem to Jericho: A study of
situational and dispositional variables in helping behavior. Journal of
Personality and Social Psychology, 27, 100–108. doi:10.1037/h0034449
Darley, J. M., & Latan
e, B. (1968). Bystander intervention in emergencies:
Diffusion of responsibility. Journal of Personality and Social Psychology,
8, 377–383. doi:10.1037/h0025589
Dillman, D. A. (2000). Mail and Internet surveys: The tailored design
method (2nd ed.). New York, NY: Wiley.
Eisenberg, N., & Spinrad, T. L. (2014). Multidimensionality of prosocial
behavior. In L. Padilla-Walker & G. Carlo (Eds.), The complexities of
raising prosocial children: An examination of the multidimensionality of
prosocial behaviors (pp. 17–39). Oxford, UK: Oxford University Press.
Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G
Power 3: A
flexible statistical power analysis program for the social, behavioral,
and biomedical sciences. Behavior Research Methods, 39, 175–191.
doi:10.3758/BF03193146
Freund, A. M., & Blanchard-Fields, F. (2014). Age-related differences in
altruism across adulthood: Making personal financial gain versus con-
tributing to the public good. Developmental Psychology, 50, 1125–1136.
doi:10.1037/a0034491
Frimer, J. A., & Walker, L. J. (2009). Reconciling the self and morality: An
empirical model of moral centrality development. Developmental Psy-
chology, 45, 1669–1681. doi:10.1037/a0017418
Garson, D. G. (2008). Factor analysis: Statnotes. Retrieved from http://
www2.chass.ncsu.edu/garson/pa765/factor.htm
Goodman, J. K., Cryder, C. E., & Cheema, A. (2013). Data collection in a
flat world: Strengths and weaknesses of Mechanical Turk samples. Jour-
nal of Behavioral Decision Making, 26, 213–224. doi:10.1002/bdm.1753
Hine, D. W., & Gifford, R. (1991). Fear appeals, individual differences and
environmental concern. Journal of Environmental Education, 23, 36–
41. doi:10.1080/00958964.1991.9943068
Jones, L. W., Sinclair, R. C., & Courneya, K. S. (2003). The effects of source
credibility and message framing on exercise intentions, behaviors, and
attitudes: An integration of the elaboration likelihood model and pros-
pect theory. Journal of Applied Social Psychology, 33, 179–196.
doi:10.1111/j.1559-1816.2003.tbh02078.x
Kahana, E., Bhatta, T., Lovegreen, L. D., Kahana, B., & Modlarksy, E. (2013).
Altruism, helping, and volunteering: Pathways to well-being inlate life. Jour-
nal of Aging and Health, 25, 59–87. doi:10.1177/0898264312469665
Kaiser, H. F. (1960). The application of electronic computers to factor
analysis. Educational and Psychological Measurement, 20, 141–151.
doi:10.1177/001316446002000116
Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1986). Fairness and the
assumptions of economics. Journal of Business, 59, S285–S300.
doi:10.1086/296367
Lee, Y.-R., & Sung, K.-T. (1997). Cultural differences in caregiving motiva-
tions for demented parents: Korean caregivers versus American care-
givers. The International Journal of Aging and Human Development,
44, 115–127. doi:10.2190/BLY5-07M1-6VVY-XJMX
Lescano, C. M. (1999). Promoting sun-protective behaviors through a
school plus home intervention. Dissertation Abstracts International:
Section B: The Sciences and Engineering, 60, 835.
Murphy, R. O., & Ackermann, K. A. (2014). Social value orientation theoretical
and measurement issues in the study of social preferences. Personality and
Social Psychology, 18, 13–41. doi:10.1177/1088868313501745
Ouellette, J. A., & Wood, W. (1998). Habit and intention in everyday
life: The multiple processes by which past behavior predicts future
behavior. Psychological Bulletin, 124, 54–74. doi:10.1037/0033-2909.
124.1.54
Pashak, T. J., & Laughter, T. C. (2012). Measuring service-mindedness and
its relationship with spirituality and life satisfaction. College Student
Journal, 46, 183–192. doi:10.1111/j.2150-1092.2012.00045_40.x
Pavey, L., Greitemeyer, T., & Sparks, P. (2011). Highlighting related-
ness promotes prosocial motives and behavior. Personality and
Social Psychology Bulletin, 37, 905–917. doi:10.1177/0146167
211405994
Pavey, L., Greitemeyer, T., & Sparks, P. (2012). I help you because I want
to, not because you tell me to: Empathy increases autonomously moti-
vated helping. Personality and Social Psychology Bulletin, 38, 681–689.
doi:10.1177/0146167211435940
Penner, L. A. (2002). Dispositional and organizational influences on sus-
tained volunteerism: An interactionist perspective. Journal of Social
Issues, 58, 447–468. doi:10.1111/1540-4560.00270
Pew Research Center. (2017). Tech adoption climbs among older adults.
Retrieved from C:/Users/Rachel/Downloads/PI_2017.05.17_Older-
Americans-Tech_FINAL.pdf
Rushton,J.P.,Chrisjohn,R.D.,&Fekken,G.C.(1981).Thealtruis-
tic personality and the self-report altruism sale. Personality and
Individual Differences, 1, 292–302. doi:10.1016/0191-8869(81)
90084-2
Sage, L., Kavussanu, M., & Duda, J. (2006). Goal orientations and moral
identity as predictors of prosocial and antisocial functioning in male
association football players. Journal of Sports Sciences, 24, 455–466.
doi:10.1080/02640410500244531.
Scales, P. S., & Benson, P. L. (2003, March). Indicators of positive youth
development: Prosocial orientation and community service. Paper pre-
pared for Indicators of Positive Youth Development Conference, Child
Trends, Washington, DC.
Siegel, J. T., Alvaro, E. A., Crano, W. D., Gonzalez, A., Tang, J., & Jones, S.
P. (2010). Passive-positive organ donor registration behavior: A mixed
method assessment of the IIFF Model. Psychology, Health and Medi-
cine, 15, 198–209. doi:10.1080/13548501003623922
Siegel, J. T., Thomson, A. L., & Navarro, M. N. (2014). Experimentally dis-
tinguishing elevation from gratitude: Oh, the morality. Journal of Positive
Psychology, 9, 414–427. doi:10.1080/17439760.2014.910825
Sirgy, M. J., Gurel-Atay, E., Webb, D., Cicic, M., Husic, M., Ekici, A., …&
Johar, J. S. (2012). Materialism Scale—Modified [Database record].
Retrieved from PsycTESTS. doi:10.1037/t32070-000
Smith, J. R., & McSweeney, A. (2007). Charitable giving: The effectiveness
of a revised theory of planned behaviour model in predicting donating
intentions and behaviour. Journal of Community & Applied Social Psy-
chology, 17, 363–386. doi:10.1002/casp.906
Szreter, S., & Woolcock, M. (2004). Health by association? Social capital,
social theory, and the political economy of public health. International
Journal of Epidemiology, 33, 650–667. doi:10.1093/ije/dyh013
Thomson, A. L., Nakamura, J. N., Siegel, J. T., & Csikszentmihalyi, M.
(2014). Elevation and mentoring: An experimental assessment of causal
relations. Journal of Positive Psychology, 9, 402–413. doi:10.1080/
17439760.2014.910824
Van Tongeren, D. R., Green, J. D., Davis, D. E., Hook, J. N., & Hulsey, T. L.
(2016). Prosociality enhances meaning in life. The Journal of Positive
Psychology, 11, 225–236. doi:10.1080/17439760.2015.1048814
PROSOCIAL BEHAVIORAL INTENTIONS SCALE 9
Webb,T.,&Sheeran,P.(2006).Doeschanging behavioral intentions
engender behavior change? A meta-analysis of experimental evi-
dence. Psychological Bulletin, 132, 249–268. doi:10.1037/0033-
2909.132.2.249
Winterich, K. P., Aquino, K., Vikas, M., & Swartz, R. (2013). When moral
identity symbolization motivates prosocial behavior: The role of recog-
nition and moral identity internalization. Journal of Applied Psychology,
98, 759–770. doi:10.1037/a0033177
Appendix
Prosocial Behavioral Intentions Scale
Instructions: Imagine that you encounter the following opportunities to help others. Please indicate how willing you would be to
perform each behavior from 1 (Definitely would not do this)to7(Definitely would do this). If you are more likely to complete one
task (e.g., help a stranger find a key) than another (e.g., help a stranger find a missing pet), please respond to the task that you
would be more likely to perform.*
1. Comfort someone I know after they experience a hardship
2. Help a stranger find something they lost, like their key or a pet
3. Help care for a sick friend or relative
4. Assist a stranger with a small task (e.g., help carry groceries, watch their things while they use the restroom)
Scoring: Calculate the mean of scores on all items.
*The final sentence was not part of the original instructions, but is recommended for future use.
10 BAUMSTEIGER AND SIEGEL
