Ask in person: You're less persuasive than you think over email

Article (PDF Available)inJournal of Experimental Social Psychology 66:223-226 · March 2017with 1,456 Reads
DOI: 10.1016/j.jesp.2016.10.002
Research has found people underestimate the likelihood strangers will comply with their direct requests (Bohns, 2016; Flynn & Lake (Bohns), 2008). Here we argue this “underestimation-of-compliance effect” may be limited to requests made face-to-face. We find when making direct requests over email, requesters instead overestimate compliance. In two studies, participants asked strangers to comply with requests either face-to-face or over email. Before making these requests, requesters estimated the number of people they expected to say “yes”. While requesters underestimated compliance in face-to-face contexts, replicating previous research, they overestimated compliance in email contexts. Analyses of several theorized mechanisms for this finding suggest that requesters, anchored on their own perspectives, fail to appreciate the suspicion, and resulting lack of empathy, with which targets view email requests from strangers. Given the prevalence of email and text-based communication, this is an extremely important moderator of the underestimation-of-compliance effect.
Figures - uploaded by Vanessa Bohns
Author content
All content in this area was uploaded by Vanessa Bohns
Ask in person: You're less persuasive than you think over email
M. Mahdi Roghanizad
, Vanessa K. Bohns
University of Waterloo, Canada
Cornell University, United States
People underestimate compliance when making requests of strangers in person.
In two studies, we found the opposite pattern of results for emailed requests.
Requesters overestimated compliance when making requests over email.
This error was driven by a perspective-taking failure.
Requesters failed to appreciate how untrustworthy their emails would seem to others.
abstractarticle info
Article history:
Received 12 May 2016
Revised 7 September 2016
Accepted 4 October 2016
Available online xxxx
Research has found people underestimate thelikelihood strangers will comply with theirdirect requests (Bohns,
2016; Flynn& Lake (Bohns), 2008). Here we argue this underestimation-of-complianceeffectmay be limitedto
requests made face-to-face. We nd when making direct requests over email, requesters instead overestimate
compliance. In two studies, participants asked strangers to comply with requests either face-to-face or over
email. Before making these requests, requesters estimated the number of people they expected to say yes.
While requesters underestimated compliance in face-to-face contexts, replicating previous research, they
overestimated compliance in email contexts. Analyses of several theorized mechanisms for this nding suggest
that requesters, anchored on their own perspectives, fail to appreciate the suspicion, and resulting lack of empa-
thy, with which targets view email requests from strangers. Given the prevalence of email and text-based com-
munication, this is an extremely important moderator of the underestimation-of-compliance effect.
© 2016 Published by Elsevier Inc.
Social inuence
1. Introduction
A growing body of research nds people are undercondent in their
ability to persuade others to comply with their requests (Bohns et al.,
2011; Bohns, Roghanizad, & Xu, 2014; Bohns, Newark, & Xu, 2016;
Flynn & Lake (Bohns), 2008; Newark, Flynn, & Bohns, 2014). Across at
least 12 studies in which participants collectively have asked over
14,000 strangers to comply with requests such as completing a ques-
tionnaire and borrowing a phone, participants appear consistently to
underestimateby a large marginthe likelihood people they approach
will say yes(Bohns, 2016).
This phenomenon results from requesters' inability to appreciate the
perspective of targets of their requests. Targets feel awkward and un-
comfortable saying no,both because of what it might insinuate
about the requester (Bohns, 2016; Sah, 2012), and because it feels bad
to let someone down (Newark, Bohns, & Flynn, 2016). However, re-
questers are anchored on their own perspectives and fail to recognize
the pressure targets feel to comply (Epley, Keysar, Van Boven, &
Gilovich, 2004). Consequently, requesters wrongly assume it is
easyand therefore likelyfor targets to say no.
Here we propose a moderator of the underestimation-of-compli-
ance effect. We theorize the tendency to underestimate compliance is
limited to face-to-face interactions. Specically, we hypothesize when
making direct requests over email, requesters will overestimate,notun-
derestimate, compliance. Given the prevalence of email and text-based
communication, this would be an extremely important moderator of
the original effect.
The hypothesis that requesters will overestimate compliance when
making requests over email follows from a theorized perspective-taking
failure similar to that underlying the original effect. However, dueto the
considerable differences between email and face-to-face communica-
tion, the specic mechanisms involvedand resulting predictionare
notably different.
Journal of Experimental Social Psychology xxx (2016) xxxxxx
Corresponding author at: University of Waterloo, 200 University Avenue, Waterloo,
ON, Canada.
E-mail address: (M.M. Roghanizad).
YJESP-03478; No. of pages: 4; 4C:
0022-1031/© 2016 Published by Elsevier Inc.
Contents lists available at ScienceDirect
Journal of Experimental Social Psychology
journal homepage:
Please cite this article as: Roghanizad, M.M., & Bohns, V.K., Ask in person: You're less persuasive than you think over email, Journal of Experimental
Social Psychology (2016),
Most relevant to the current research of the many ways in which
email differs from face-to-face communication is its restriction of non-
verbal cues that generate trust and empathy. Essentially, it is easier for
an unfamiliar requester to appear well-meaning and sympathetic
face-to-face than over email (Berry & McArthur, 1986; Brownlow,
1992; Burgoon, 1990; McGinley, LeFevre, & McGinley, 1975;
Scharlemann, Eckel, Kacelnik, & Wilson, 2001; Sproull & Kiesler, 1986;
Willis & Todorov, 2006). Indeed, requests made face-to-face are far
more effective than those made otherwise (Constant, Sproull, &
Kiesler, 1996; Dabbish, Kraut, Fussell & Kiesler, 2005; Gerber & Green,
2000; Ling et al., 2005; Zhu et al., 2016).
Yet requesters likely do not recognize the effect of these limitations
of email. Anchored on the intimate knowledge they have of their own
trustworthiness and circumstances, we theorize requesters will struggle
to envision what their targets see: a suspicious email from a stranger
that generates little empathy. This error should lead requesters to over-
estimate compliance over email.
We tested this hypothesis in two studies in which participants made
actual requests of strangers face-to-face or over email after predicting
the likelihood targets would comply. For both studies, we report all
measures, conditions, data exclusions, and how we determined sample
2. Study 1
506 university students participated (49 requesters, 457 targets).
Four requesters in the face-to-face condition did not complete the
study, leaving 495 participants (45 requesters [31 female], 450 targets).
Requesters received $10; targets receivedno compensation. Sample size
was determined by the sample size used by Flynn and Lake (Bohns)
(2008);Study 1;N= 23 requesters) who originally identied the un-
derestimation-of-compliance effect and whose paradigm we adapted.
The original effect was large (d= 1.096), so this sample size ensured
N80% power.
Requesters were randomly assigned to face-to-faceor emailcon-
ditions and instructed to ask 10 strangersto complete a questionnaire (a
44-item personality inventory; John, Donahue, & Kentle, 1991). First, re-
questers were provided with complete information about their task; no
information was withheld. They then predicted how many of the 10
people they approached/emailed would comply with their requests. Be-
cause of the role of discomfort in previous research, we also adminis-
tered a measure of predicted discomfort (Appendix A).
Next, requesters in the face-to-face condition went onto the univer-
sity campus with a stackof questionnairesand a tally sheet to recordthe
responses of the 10 strangers they approached. In the email condition,
requesters were given 10 email addresses from the university directory
and asked if they recognized any addresses. They then sent emails with
the request to complete an online questionnaire one at a time using
their own university email accounts.
We recorded actual compliance
in this condition through the online questionnaire.
To ensure the face-to-face and email conditions were comparable in
all respects aside from communication medium, thescripts participants
used when making their requests were written to be as similar as possi-
ble, while also conveying the same information in the email that would
be implicitly conveyed in a face-to-face interaction on campusnamely,
that the requester was a student asking a fellow student (Appendix A).
To determine whether participants in the face-to-face condition
were strategically approaching participants in ways that would be im-
possible in the email condition, we analyzed targets' gender composi-
tion. Requesters may have approached slightly more female targets
(54%) than would be expected had they been approaching targets ran-
domly compared to the general campus population (45% female),
t(25) = 3.52, p= 0.002, but not compared to the population of the
social sciences campus where the study took place (60% female),
t(25) = 1.73, p= 0.095 (CUDO, 2015). Importantly, female targets
were no more likely to comply (71.1%) than male targets (72.9%),
(N= 260) = 0.098, p= 0.75, so any employed strategy was
2.1. Results
A 2 × 2 mixed-model ANOVA with repeated measures on the second
factor revealed an interaction between Request Medium (face-to-face,
email) and compliance (predicted, actual), F(1,43) = 121.10,
pb0.001, η
=0.73(Fig. 1).
Requesters underestimated the likelihood
targets would comply with their requests face-to-face (Predicted: M=
5.08, SD = 2.23; Actual: M=7.15,SD = 1.81), F(1,25) = 17.45,
pb0.001, d= 1.00), replicating previous research. However, requesters
overestimated the likelihood targets would comply with their emailed
requests (Predicted: M=5.53,SD = 1.71; Actual: M=0.21,SD =
0.54), F(1,18) = 185.47, pb0.001, d= 4.20). Although targets asked
to complete a questionnaire face-to-face were 34 times more likely to
comply than those asked over email, F(1,44) = 260.78, pb0.001, d=
5.20, requesters' predictions of compliance did not differ between the
conditions, F(1,44) = 0.53, pN0.250, d= 0.21.
Notably, requesters recognized it would be more difcult for targets
to say noface-to-face M= 3.14 (SD = 1.08) than over email (M=
1.75, SD =0.87),F(1,44) = 21.47, pb0.001, d= 1.42, suggesting that
inaccurate assumptions about the discomfort of saying nowere not
driving our main nding.
2.2. Discussion
Despite nding a large effect of Request Medium on the direction of
requesters' prediction error, we were unable to identify the mechanism
underlying this reversal of the original effect. Further, the paradigm we
used only allowed us to capture requesters', not targets', responses to
our mechanism questions. Notably, collecting the latter measures
would require questionnaire data bothfrom targets who agreed to com-
plete a questionnaire and those who refused to complete a question-
naire. In Study 2, we used a unique study design to collect this data.
We also rened our procedure to address the possibility that partici-
pants in the face-to-face condition approached targets strategically. Fi-
nally, we included items to test the mechanism proposed
earliernamely, that requesters fail to appreciate the implicit trust
granted in face-to-face interactions, but not over email, which leads to
Towards the end of Study 1, our university ethics board became concerned that uni-
versity students were spamming other students. We were asked to stop collecting data
and reassess the email condition of the study. Thus, there is a smaller number of partici-
pants in the email condition than the face-to-face condition, and some changes to our
method were imposed in Study 2.
FtF email
Fig. 1. Actual vs. predictedcompliance in the face-to-faceand email conditionsin Study 1.
Fluctuations in degrees of freedomwithin a studyoccur when a participant did notan-
swer a question, thus reducing the sample size for that question.
2M.M. Roghanizad, V.K. Bohns / Journal of Experimental Social Psychology xxx (2016) xxxxxx
Please cite this article as: Roghanizad, M.M., & Bohns, V.K., Ask in person: You're less persuasive than you think over email, Journal of Experimental
Social Psychology (2016),
increased empathy towards the requester, and ultimately higher rates
of compliance.
3. Study 2
480 university students participated (60 requesters [36 female] and
420 targets). Requesters received $10; targets received $1. Sample size
was determined by the sample size of Flynn and Lake (Bohns) (2008),
again ensuring N80% power.
Requesters were randomly assigned to face-to-faceor emailcon-
ditions. In order to collect mechanism data both from targets that said
yesand those that said noto completing a questionnaire, requesters
were instructed to ask strangers who had already agreed to ll out a
one-page questionnaire for $1 to complete an additional task (editing
a document for grammatical mistakes) for free. Before making these re-
quests, requesters predicted the number of people (out of 7) who would
agree to complete thefree task and answered our mechanism questions.
Requesters were again provided with complete taskdetails before com-
pleting these measures.
In the face-to-face condition, requesters approached as many
strangers as necessary to recruit 7 people to ll out a questionnaire for
$1. When someone agreed to complete the paid questionnaire, re-
questers immediately asked the target to complete the additional task
for no additional pay. To ensure requesters were approaching targets
randomly, requesters were instructed to count six strangers and ap-
proach the sixth.
In the email condition, emails were sent to 7 university students
who had previously registered to complete a questionnaire for $1. The
email text mirrored the face-to-face script. Email recipients who agreed
to complete a questionnaire for $1 were asked if they would complete
the extra task for free. As requested by our university ethics board, to
protect participants' privacy requesters no longer sent emails individu-
ally to targets after completing the initial materials. Instead, a university
technician sent a batch of 210 emails (7 for each of the30 participants in
the email condition) from a ctitious university email address. Howev-
er, to increase perceptions of the authenticity of their requests, re-
questers were provided with 7 bogus email addresses ostensibly from
our pre-registered list of targets and were asked if they recognized
any addresses.
Targets in both conditions who agreed to complete the paid ques-
tionnaire said either yesor noto completing the unpaid task. They
then completed the paid questionnaire, which comprised a series of
questions about why they had said yesor noto completing the un-
paid task. Thus, we have mechanism data from (a) requesters, (b) tar-
gets who said yesto the unpaid task, and (c) targets who said no
to the unpaid task. This data includes how uncomfortable targets felt/
would feel saying noto the unpaid task, how much targets trusted/
would trust requesters, and how much targets empathized/would em-
pathize with requesters (Appendix A).
3.1. Results
3.1.1. Overview
Although 210 pre-registered targets received emails, only 44 com-
pleted the paid questionnaire. For presentation clarity, we created a
simulated sample of 210 targets for the email condition by drawing 30
samples of 7 respondents with replacement from the 44 targets who
completed the paid questionnaire. The conclusions drawn from this
simulated sample match those of the raw sample (see Appendix B for
raw data analyses).
To be consistent with the analyses in Study 1, we modeled the data
at the requester level (N= 60). We assigned to each requester an actu-
al compliancevalue comprising the total number of targets (out of 7)
who completed the free task, as well as actualtrust, empathy, and dis-
comfort values comprising averages of those 7 targets' responses to the
mechanism items. We did not conduct multi-level analyses because
requesters' predictions of compliance are continuous and the compli-
ance measure for each individual target is necessarily binary (i.e.,
yesor no).
3.1.2. ANOVAs
A 2 × 2 mixed-model ANOVA with repeated measures on the second
factor revealed a signicant interaction between Request Medium
(face-to-face, email) and compliance with thefree task (predicted, actu-
al), F(1,58) = 16.78, pb0.001, η
=0.22(Fig. 2). Requesters in the face-
to-face condition underestimated the likelihood targets would comply
(Predicted: M=4.43,SD = 1.70; Actual: M= 5.43, SD =1.81;
F(1,29) = 4.49, p= 0.043, d= 0.57), while requesters in the email con-
dition overestimated the likelihood targets would comply (Predicted:
M= 4.10, SD = 2.01; Actual: M= 2.43, SD = 1.31; F(1,29) = 18.84,
pb0.001, d= 0.98). Targets were again much more likely to comply
in the face-to-face condition, F(1,58) = 54.10, pb0.001, d= 1.90, but
there was no difference between requesters' predictions of compliance
between the two conditions, F(1,58) = 0.48, pN0.250, d=0.18.
Additional mixed-model ANOVAs on our mechanism indices re-
vealed a signicant interaction on discomfort, F(1,58) = 5.79, p=
0.019, η
= 0.09. Requesters recognized targets would feel more un-
comfortable saying noface-to-face (M= 3.59, SD = 1.11) than over
email (M= 2.49, SD = 1.15), F(1,58) = 14.38, p=b0.001, d= 0.97.
Targets conrmed this intuition, reporting they would feel more un-
comfortable saying noface-to-face (M= 3.35, SD = 0.43) than over
email (M= 2.98, SD =0.47),F(1,58) = 10.07, p= 0.002, d=0.82.
Thus, this mechanism again failed to explain our main nding.
However, there was a signicant interaction on trust, F(1,58) =
8.997, p= 0.004, η
= 0.134, which mirrored our compliance results.
Targets trusted requesters more in the face-to-face condition (M=
5.66, SD = 0.43) than the email condition (M=4.37,SD = 0.47),
F(1,58) = 120.98, pb0.001, d= 2.86, although requesters predicted
no difference between the two conditions, F(1,58) = 2.36, p= 0.130,
d= 0.397.
A similar (though non-signicant) interaction to that for trust
emerged for empathy, F(1,58) = 2.35, p= 0.131, η
= 0.039. Targets
felt more empathy towards requesters in the face-to-face condition
(M=4.63,SD = 0.56) than the email condition (M=3.96,SD =
0.46), F(1,58) = 25.899, pb0.001, d= 1.31. However, requesters pre-
dicted no difference between the two conditions, F(1,58) = 0.17,
pN0.250, d=0.106.
3.1.3. Mediation
Mediation analysis conrmed that the error in requesters' predic-
tions of the amount of trust and empathy targets would feel in response
to emailed versus in-person requests explained the error in requesters'
predictions of compliance between the two conditions. That is, the the-
orized path of request medium predicted minus actual trust predicted
FtF email
Fig. 2. Actual vs. predictedcompliance in the face-to-faceand email conditionsin Study 2.
3M.M. Roghanizad, V.K. Bohns / Journal of Experimental Social Psychology xxx (2016) xxxxxx
Please cite this article as: Roghanizad, M.M., & Bohns, V.K., Ask in person: You're less persuasive than you think over email, Journal of Experimental
Social Psychology (2016),
minus actual empathy compliance prediction error was signicant [CI:
0.0378, 0.0035] (see Appendix C for complete details).
4. Discussion
Overall, we nd people are less inuential than they think over
email. Although requesters underestimated the likelihood people
would comply with their requests in person, they overestimated the
likelihood people would comply with their requests over email. These
ndings appear to result from requesters' failure to appreciate the im-
plicit trust conveyed in face-to-face interactions and lost over email,
which activates targets' empathy towards requesters.
These ndings contribute to a burgeoning area of research on
people's perceptions of their inuence over others (Bohns, 2016).
While much research on social inuence aims to identify effective inu-
ence techniques, we examine people's assumptions about how effective
various inuence tactics are likely to be.
This work also contributes a new perspective to a growing body of
literature on trust in computer-mediated interactions (John, Acquisti,
& Loewenstein, 2011; Roghanizad & Neufeld, 2015). Rather than focus-
ing on users' willingness to trust computer-mediated content, our re-
search has implications for how the creators of such content are likely
to view its trustworthiness.
Practically, as computer-mediated communication becomes the
dominant means of interacting with others (Dimmick, Kline, &
Stafford, 2000), these ndings suggest that users may not realize its lim-
itations (cf., Kruger, Epley, Parker, & Ng, 2005). It is often more conve-
nient and comfortable to make requests over email. If people also
overestimate email's effectiveness, they may choose inferior means of
inuence without recognizing the downsides. Reecting on the experi-
ence of receiving an email from a stranger before making a request in
this manner may facilitate the practice of asking in person when
Strengths of these ndings include the fact that participants made
actual requests of other people in these studies, and the large effects,
ranging from d= 0.98 to d= 4.20 for the overestimation-of-compli-
ance effect over email and d= 0.57 to d= 1.01 for the underestima-
tion-of-compliance effect face-to-face. However, there are noteworthy
limitations. For one, the mechanism identied here differs from previ-
ous research on the underestimation-of-compliance effect. Thus,our ex-
planation for these ndings is still somewhat exploratory, necessitating
replication by future studies. However, it is certainly possible that errors
related to predicting compliance are multiply determined. Further, our
ndings arelimited to the specic university population and types of re-
quests we used in the current studies, as well as a paradigm in which re-
questers evaluated the effectiveness of email and face-to-face requests
separately, rather than comparing them directly. Future research should
explore the generalizability of these ndings.
Supplementary data to this article can be found online at http://dx.
The studies reported in this article are from the rst author's disser-
tation. This research was supported by funding from the Social Sciences
and Humanities Research Council of Canada (435-0012-2014) to the
second author.
Berry, D. S. D.,& McArthur,L. Z. L. (1986). Perceivingcharacter in faces: The impact of age-
related craniofacial changes on social perception. Psychological Bulletin,100(1) (3-3-
Bohns, V. K. (2016). (Mis)understanding our inuence over others: A review of the un-
derestimation-of-compliance effect. Current Directions in Psychological Science,25,
Bohns, V. K., Handgraaf, M. J. J., Sun, J., Aaldering, H., Mao, C., & Logg, J. (2011). Are social
prediction errors universal? Predicting compliance with a direct request across cul-
tures. Journal of Experimental Social Psychology,47(3), 676680.
Bohns, V. K., Roghanizad, M., & Xu, A. Z. (2014). Underestimating our inuence over
others' unethical behavior and decisions. Personality and Social Psychology Bulletin,
40(3), 348362.
Bohns, V. K., Newark, D., & Xu, A. (2016). For a dollar, would you? How (we think)
money inuences c ompliance with ou r requests. Orga nizational Behavior and
Human Decision Processes,134,4562.
Brownlow, S. (1992). Seeing is believing: Facia l appearance, credibi lity, and attitude
change. Journal of Nonverbal Behavior,16(2), 101115.
Burgoon, J. K. (1990). Nonverbal behavior s, persuasion, and credibility . Human
Communication Research,17(1), 140169.
Constant, D., Sproull, L., & Kiesler, S. (1996). The kindness of strangers: The usefulness of
electronic weak ties for technical advice. Organization Science,7(2), 119135.
CUDO (Common University Data Ontario) (2015). A-4 (Female Enrolment by Program)
and A-5 (Male Enrolment by Program).
Dabbish, L. A., Kraut, R. E., Fussell, S., & Kiesler, S. (2005, April). Understanding email use:
predicting action on a message. In Proceedingsof the SIGCHI conference on Human fac-
tors in computing systems. ACM (pp. 691700).
Dimmick, J., Kline, S., & Stafford, L. (2000). The gratication niches of personal e-mail and
the telephone competition, displacement, and comple mentarity. Communication
Research,27(2), 227248.
Epley, N., Keysar, B., Van Boven, L., & Gilovich, T. (2004). Perspective taking as egocentric
anchoring and ad justment. Journal of Personality a nd Social Psycholo gy,87(3),
Flynn, F. J., & Lake (Bohns), V. K. B. (2008). If you need help, just ask: Underestimating
compliance with direct requests for help. Journal of Personality and Social
Psychology,95(1), 128143.
Gerber, A. S., & Green, D. P. (2000). The effects of canvassing, telephone calls, and direct
mail on voter turnout: A eld experiment. American Political Science Review,94(03),
John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The big ve inventoryVersions 4a and
54. Berkeley, CA: University of California, Berkeley, Institute of Personality and Social
John, L. K., Acquisti, A., & Loewenstein, G. (2011). Strangers on a plane: Context-depen-
dent willingness to divulge sensitive information. Journal of Consumer Research,
37(5), 858.
Kruger, J. J., Epley, N. N., Parker, J. J., & Ng, Z. Z. (2005). Egocentrism over e-mail: Can we
communicate as well as we think? Journal of Personality and Social Psychology,89(6),
Ling, K., Beenen, G., Ludford, P., Wang, X., Chang, K., Li, X., & Resnick, P. (2005). Using so-
cial psychology to mo tivate contribu tions to online communities. Journal of
ComputerMediated Communication,10(4) 0000.
McGinley, H., LeFevre, R., & McGinley, P. (1975). The inuence of a communicator's body
position on opinion change in others. Journal of Personality and Social Psychology,
31(4), 686690.
Newark, D.A., Flynn, F. J., & Bohns,V. K. (2014). Once bitten, twice shy:The effect of a past
refusal on expectations of future complia nce. Social Psycholo gical and Personality
Science,5(2), 218225.
Newark, D. A., Bohns, V. K., & Flynn, F. J. (2016). The value of a helping hand: Help-seekers'
predictions of help quality. Manuscript submitted for publication.
Roghanizad, M. M., & Neufeld, D. J. (2015). Intuition, risk, and the formation of online
trust. Computers in Human Behavior,50, 489498.
2015.04.025 (Complete).
Sah, S. (2012).Conicts of interestand your physician:Psychological processes thatcause
unexpected changes in behavior. The Journal of L aw, Medicine & Et hics,40(3),
Scharlemann, J. P. W.,Eckel, C. C., Kacelnik, A., & Wilson, R. K. (2001). Thevalue of a smile:
Game theory with a human face. Journal of Economic Psychology,22(5), 617.
Sproull, L., & Kiesler, S. (1986). Reducing social context cues: Electronic mail in organiza-
tional communication. Management Science,32(11), 1492.
Willis, J ., & Todorov, A . (2006). First impressions: Makingup your mind after a 100-ms ex-
posure to a face. Psychological Science (Wiley-Blackwell),17(7), 592598.
Zhu, H., Das, S., Cao, Y., Yu, S., Kittur, A., & Kraut, R. (2016, May). A Market in Your Social
Network: The Effects of Extrinsic Rewards on Friendsourcing and Relationships. In
Proceedings of th e 2016 CHI Conferen ce on Human Factors in Computing Systems
(pp. 598609). ACM.
4M.M. Roghanizad, V.K. Bohns / Journal of Experimental Social Psychology xxx (2016) xxxxxx
Please cite this article as: Roghanizad, M.M., & Bohns, V.K., Ask in person: You're less persuasive than you think over email, Journal of Experimental
Social Psychology (2016),
  • ... Some examples of compliance problems include decisions related to buying a product, donating to a charity, or, generally, complying with a request. The literature in psychology, communications, and political science has noted the face effect, i.e., face-to-face interactions are more effective in inducing compliance than other forms of interactions, such as direct mail, telephone calls, and emails (e.g., Milgram, 1965b;Gerber and Green, 2000;Roghanizad and Bohns, 2017). ...
    ... More recent research has noted the face effect (e.g., Milgram, 1965b;Gerber and Green, 2000;Roghanizad and Bohns, 2017). For example, Gerber and Green (2000) found that personal canvassing increases voter turnout more than direct mail and telephone calls in a field experiment. ...
    ... For example, Gerber and Green (2000) found that personal canvassing increases voter turnout more than direct mail and telephone calls in a field experiment. Roghanizad and Bohns (2017) suggest that people often underestimate compliance rate in face-to-face interactions while overestimating compliance rate of emails due to varied trust and empathy levels of the two channels. Yet the underlying mechanism of the face effect has rarely been explicitly explored (Gerber and Green, 2000). ...
    Full-text available
    Face-to-face interactions are central to many individual choices and decision-making issues, such as customer services, sales, promotions, and negotiations. While the face effect, that is, face-to-face interactions are more effective in inducing compliance than other forms of interactions, has been noted in the literature, its mechanism has rarely been explored. This research helps to fill the theoretical void and provides new insights into the face effect with two lab experiments and one field experiment. Study 1, a field experiment conducted in a beauty salon, and Study 2, a lab experiment, show that the face effect is largely attributable to anticipated facial feedback and that the face effect is stronger when individuals are sensitive to face and when the requester’s face is expressive. Study 3, using video-simulated face-to-face interactions, demonstrates that anticipated facial feedback, not necessarily actual feedback, is enough to drive the face effect. In so doing, this research furthers our understanding of factors that affect individual compliance in face-to-face interactions in both the “sending” and “receiving” stages. We discuss the theoretical and empirical implications, limitations, and future avenues of research.
  • ... In these studies, participants consistently and substantially under- estimated the likelihood that others would comply with their direct, face-to-face requests, because they underestimated the power of the strong social forces that drive people to say "yes" (by making it awkward and uncomfortable to say "no") in these situations. These findings have since been replicated and extended in numerous follow-up studies (Bohns et al., 2011(Bohns et al., , 2016Newark, Bohns, & Flynn, 2017;Newark, Flynn, & Bohns, 2014;Roghanizad & Bohns, 2017). This underestimation-of-compliance effect, like the invisibility cloak, would predict an overall tendency to underestimate one's influence over others. ...
    Full-text available
    We explore how, and how accurately, people assess their influence over others’ behavior and attitudes. We describe the process by which a person would determine whether he or she was responsible for changing someone else’s behavior or attitude, and the perceptual, motivational, and cognitive factors that are likely to impact whether an influencer’s claims of responsibility are excessive, insufficient, or accurate.
  • ... This occurred because more Calgary patients received completion re- quests in person during subsequent follow-up visits; whereas more Edmonton patients were followed-up over email. This finding is consistent with previous studies that found face-to-face requests to be more successful than email requests [48]. A future sensitivity analysis could be performed to further examine the impact of this difference; however, bias resulting from non-response in surveys is difficult to assess since information about non-responders is rarely available. ...
    Full-text available
    Background: Measuring quality in healthcare is vital in evaluating patient outcomes and system performance. The availability of reliable and valid information about the quality of care for patients presenting with rotator cuff disorders (RCD) in Alberta, Canada is scarce. The objective of this study is to measure quality of care for patients with RCD in order to identify areas of improvement. Methods: This study employs descriptive survey research design. Between March 2015 and November 2016, a convenience sample of patients presenting with chronic, full-thickness rotator cuff tears to two sport medicine centres in Calgary and Edmonton, Alberta completed two questionnaires: the Healthcare Access and Patient Satisfaction Questionnaire (HAPSQ) and the Rotator Cuff Quality-of-Life Index (RC-QOL). Data collected using both questionnaires were used to make judgments about quality of care. Quality of care was evaluated using six dimensions of quality defined by the Alberta Quality Matrix for Health: accessibility, acceptability, efficiency, effectiveness, appropriateness, and safety. Data was also used to compare current patient clinical pathways to ideal clinical pathway algorithms and used to make judgments about the appropriateness and safety of healthcare practices. Results: One hundred seventy-one patients participated in the study. The longest mean waiting times for medical services in Alberta were for magnetic resonance imaging (MRI) received in the public sector (103 days) and consultation by orthopaedic surgeon (172 days). Patient satisfaction with respect to quality of care was lowest for emergency room physician and highest for orthopaedic surgeon visits. Patients were treated by a mean of 2.5 physicians (SD: 0.77, range: 2-7). The total aggregate average cost per patient was $4541.19. The mean RC-QOL score for all patients was 42 (SD: 22). Only 54 patients (64%) requiring surgery were able to consult with a surgeon within benchmark timeframes. A comparison of current to ideal clinical pathway algorithms found that 38 patients (22%) experienced indirect clinical pathways, whereby care was fragmented and patients received care from multiple and often, redundant healthcare professionals. Conclusion: There is a discrepancy between current and ideal clinical pathways whereby some patients are experiencing quality of care that is inefficient, disjointed, and less than ideal.
  • ... Communicators tend to overestimate the degree to which recipients will comply with email (vs. face-to-face) requests, as well as the degree to which email requests will seem trustworthy ( Roghanizad & Bohns, 2017). Relatedly, the potential for misunderstanding is higher in mediated communication ( Waterworth & Waterworth, 2006). ...
    Precision medicine (PM) draws upon individual biological and psychosocial factors to create a personalized approach to healthcare. To date, little is known about how healthcare consumers will respond to such highly personalized guidance and treatment. The assumption is that responses will generally be favorable; yet in the media and in online public discussions about PM, concerns have been raised about invasions of privacy and autonomy. Findings from the tailoring literature—relevant because PM is, in a sense, “hypertailoring”—similarly suggest a potential for provoking unintended consequences such as personalization reactance, wherein perceived threat to one’s privacy or freedom can lead to rejection of the personalized message or its source. Here, we review extant tailoring and other relevant research to identify challenges that could arise in PM communication. We then draw upon a patient-centered communication perspective to highlight elements of the communication process wherein resistance could be mitigated. This review aims to provide preliminary guidance for practitioners when communicating with patients and healthcare consumers about PM, as well as point scholars toward fruitful topics for research in this emerging health communication area.
  • ... A similar parallel can be drawn with the optimization of dislocated work processes, where communication between individuals is key for the realization of a set goal. Here we must emphasize, that direct graphically conditioned communication is significantly more effective than the exchange of written messages using email or chatrooms [11]. Virtual reality thus presents itself as an effective alternative, mostly due to the cognitive perception of messages, which is similar to real relationships [12]. ...
    This paper takes the form of a discussion article that briefly analyses the societal and technological progress of individual and group interactions within virtual reality space. From basic interactions between simplified avatars in the early 1990s to more complex ones in recent years, we are witnessing the drastic popularization of virtual reality as a social tool-or social VR for short. In this article, we will take a look at early virtual reality environments, where individuals could interact both with modular objects and other individuals. To better understand recent trends, we will compare the aforementioned with modern virtual spaces and discuss their use. We will also set two examples, which point towards potential merging of the now fragmented and separated virtual environments in the near future and, in doing so, discuss the future pathways of social VR.
  • ... Podobno vzporednico bi lahko potegnili z optimizacijo dislociranih delovnih procesov, kjer je komunikacija med posamezniki ključna za realizacijo izbranega cilja. Ob tem poudarimo, da je neposredna grafično pogojena komunikacija bistveno učinkovitejša od pisne izmenjave sporočil s pomočjo spletne pošte ali klepetalnice[11]. Navidezna resničnost se tako kaže kot učinkovita alternativa predvsem zaradi kognitivne percepcije prejetih sporočil, ki je podobna stvarnim odnosom[12]. ...
    This paper overtakes the form of a discussion article that briefly analyses the societal and technological progress of individual and group interactions within virtual reality space. From basic interactions between simplified avatars in early 90. to more complex ones in recent year, we are witnessing the drastic popularization of virtual reality as social tool or social VR for short. In this article, we will take a look over early virtual reality environments where individuals could interact both with modular objects and other individuals. For the purpose of deeper understanding of recent trends, we will compare them with modern virtual spaces and discuss their use. We will also set two examples that point towards potential merging of now fragmented and separated virtual environments in the near future and by that discuss the future pathways of social VR.
  • Article
    We present a theoretical model that describes the interaction between social and human capital and the different forms that such an interaction may take – positive and negative. Extending the concepts of stocks and flows of knowledge, this model evaluates social capital flows against human capital stocks. When we compare these flows to the value of human capital, we are able to better understand how these resources can interact with each other, influence global talent development, and change over time. We discuss the implications for practices in global talent management in selecting, developing, and harnessing talent.
  • Article
    We extend research on charity donations by exploring an everyday tactic for increasing compliance: asking politely. We consider three possible effects of politeness on charity donations: a positive effect, a negative effect, and a wiggle‐room effect where the perception of the request is adjusted to decline donating without feeling selfish. Results from six experiments systematically supported the polite wiggle‐room effect. In hypothetical donations contexts, indirect requests were judged more polite. In real donation contexts, though, indirect requests were not judged as more polite and had no consistent effect on donation decision. Rather, the decision to donate predicted the perceived politeness of the request, independently of its phrasing. Experiment 4 provided causal evidence that participants justified their donation decisions by adjusting their perception of the request. The polite wiggle‐room effect has important implications for organizations that seek to increase compliance while maintaining a positive image.
  • Conference Paper
    Friendsourcing consists of broadcasting questions and help requests to friends on social networking sites. Despite its potential value, friendsourcing requests often fall on deaf ears. One way to improve response rates and motivate friends to undertake more effortful tasks may be to offer extrinsic rewards, such as money or a gift, for responding to friendsourcing requests. However, past research suggests that these extrinsic rewards can have unintended consequences, including undermining intrinsic motivations and undercutting the relationship between people. To explore the effects of extrinsic reward on friends' response rate and perceived relationship, we conducted an experiment on a new friendsourcing platform - Mobilyzr. Results indicate that large extrinsic rewards increase friends' response rates without reducing the relationship strength between friends. Additionally, the extrinsic rewards allow requesters to explain away the failure of friendsourcing requests and thus preserve their perceptions of relationship ties with friends.
  • Article
    Full-text available
    Research has shown a robust tendency for people to underestimate their ability to get others to comply with their requests. In five studies, we demonstrate that this underestimation-of-compliance effect is reduced when requesters offer money in exchange for compliance. In Studies 1 and 2, participants assigned to no-incentive or monetary-incentive conditions made actual requests of others. In both studies, requesters who offered no incentives underestimated the likelihood that those they approached would grant their requests; however, when requesters offered monetary incentives, this prediction error was mitigated. In Studies 3-5, we present evidence in support of a model to explain the underlying mechanism for this attenuation effect. Studies 3 and 4 demonstrate that offering monetary incentives activates a money-market frame. In Study 5, we find that this activation reduces the discomfort associated with asking, allowing requesters to more accurately assess the size of their request and, consequently, the likelihood of compliance.
  • Article
    Full-text available
    I review a burgeoning program of research examining people’s perceptions of their influence over others. This research demonstrates that people are overly pessimistic about their ability to get others to comply with their requests. Participants in our studies have asked more than 14,000 strangers a variety of requests. We find that participants underestimate the likelihood that the people they approach will comply with their requests. This error is robust (it persists across various samples and requests) and substantial (on average, requesters underestimate compliance by 48%). We find that this error results from requesters’ failure to appreciate the awkwardness of saying “no” to a request. In addition to reviewing evidence for the underestimation-of-compliance effect and its underlying mechanism, I discuss some factors that have been found to strengthen, attenuate, and reverse the effect. This research offers a starting point for examining a neglected perspective in influence research: the psychological perspective of the influence source.
  • Article
    Understanding how consumers evaluate website trustworthiness is a critical factor for online vendors. The dominant view espouses a deliberative trust formation process whereby shoppers evaluate security certificates, return policies, user feedback and the like, implying a highly rational underlying trust calculus. In this paper we use a laboratory experiment to explore an alternative perspective, based on the non-rational associative reasoning approach. Our findings show that when faced with a no-risk hypothetical decision about whether or not they would purchase a book from an online bookseller, subjects’ decision-making processes were indeed consistent with the dominant deliberative view. However, when confronted with a decision entailing risk (i.e., sharing sensitive personal information with an unknown website), subjects became reliant on their non-rational, gut-level intuition. We adopt a dual-process reasoning theory to make sense of these findings, and recommend that vendors take into account associative reasoning factors when designing online interfaces. Future research directions are provided.
  • Article
    Full-text available
    New marketing paradigms that exploit the capabilities for data collection, aggregation, and dissemination introduced by the Internet provide benefits to consumers but also pose real or perceived privacy hazards. In four experiments, we seek to understand consumer decisions to reveal or withhold information and the relationship between such decisions and objective hazards posed by information revelation. Our central thesis, and a central finding of all four experiments, is that disclosure of private information is responsive to environmental cues that bear little connection, or are even inversely related, to objective hazards. We address underlying processes and rule out alternative explanations by eliciting subjective judgments of the sensitivity of inquiries (experiment 3) and by showing that the effect of cues diminishes if privacy concern is activated at the outset of the experiment (experiment 4). This research highlights consumer vulnerabilities in navigating increasingly complex privacy issues introduced by new information technologies.
  • Article
    Full-text available
    We examined the psychology of "instigators," people who surround an unethical act and influence the wrongdoer (the "actor") without directly committing the act themselves. In four studies, we found that instigators of unethical acts underestimated their influence over actors. In Studies 1 and 2, university students enlisted other students to commit a "white lie" (Study 1) or commit a small act of vandalism (Study 2) after making predictions about how easy it would be to get their fellow students to do so. In Studies 3 and 4, online samples of participants responded to hypothetical vignettes, for example, about buying children alcohol and taking office supplies home for personal use. In all four studies, instigators failed to recognize the social pressure they levied on actors through simple unethical suggestions, that is, the discomfort actors would experience by making a decision that was inconsistent with the instigator's suggestion.
  • Article
    Full-text available
    The theory of the niche predicts that a new medium will compete with established media for consumer satisfaction, consumer time, and / or consumer advertising dollars. Competition between e-mail and telephone use was measured in this study at the level of gratifications derived by consumers. Gratifications and gratification opportunities (consumers' beliefs that a medium allows them to obtain greater opportunities for satisfaction) were derived from an analysis of open-ended questions. A second sample was interviewed by telephone and rated both mediums on gratification and gratification-opportunity scales. Forty-eight percent of respondents reported using the phone less since they adopted e-mail. Results indicate that a wider spectrum of needs is being served by the telephone, whereas e-mail provides greater gratification opportunities. The results indicate competition, but also indicate that the two mediums are not close substitutes.
  • Article
    Full-text available
    This study examined the relationships among nonverbal behaviors, dimensions of source credibility, and speaker persuasiveness in a public speaking context. Relevant nonverbal literature was organized according to a Brunswikian lens model. Nonverbal behavioral composites, grouped according to their likely proximal percepts, were hypothesized to significantly affect both credibility and persuasiveness. A sample of 60 speakers gave videotaped speeches that were judged on credibility and persuasiveness by classmates. Pairs of trained raters coded 22 vocalic, kinesic, and proxemic nonverbal behaviors evidenced in the tapes. Results confirmed numerous associations between nonverbal behaviors and attributions of credibility and persuasiveness. Greater perceived competence and composure were associated with greater vocal and facial pleasantness, with greater facial expressiveness contributing to competence perceptions. Greater sociability was associated with more kinesic/proxemic immediacy, dominance, and relaxation and with vocal pleasantness. Most of these same cues also enhanced character judgments. No cues were related to dynamism judgments. Greater perceived persuasiveness correlated with greater vocal pleasantness (especially fluency and pitch variety), kinesic/proxemic immediacy, facial expressiveness, and kinesic relaxation (especially high random movement but little tension). All five dimensions of credibility related to persuasiveness. Advantages of analyzing nonverbal cues according to proximal percepts are discussed.
  • Article
    Full-text available
    People use weak ties-relationships with acquaintances or strangers-to seek help unavailable from friends or colleagues. Yet in the absence of personal relationships or the expectation of direct reciprocity, help from weak ties might not be forthcoming or could be of low quality. We examined the practice of distant employees (strangers) exchanging technical advice through a large organizational computer network. A survey of advice seekers and those who replied was conducted to test hypotheses about the viability and usefulness of such electronic weak tie exchanges. Theories of organizational motivation suggest that positive regard for the larger organization can substitute for direct incentives or personal relationships in motivating people to help others. Theories of weak ties suggest that the usefulness of this help may depend on the number of ties, the diversity of ties, or the resources of help providers. We hypothesized that, in an organizational context, the firm-specific resources and organizational motivation of people who provide advice will predict the usefulness of advice. We investigated these theories in a study of employees of a global computer manufacturer. We collected survey and observational data on the relationships between information seekers and information providers; the number, diversity, resources, and motivations of information providers, and subjective ratings of the usefulness of the advice (from both parties in the exchange) and whether or not the advice solved information seekers' problems. We found that information providers gave useful advice and solved the problems of information seekers, despite their lack of a personal connection with the seekers. The data support the main hypotheses and provide some support for resource and diversity explanations of weak tie influence. We discuss how this organization's culture sustained useful information exchange through weak ties.
  • Article
    Full-text available
    Four studies examined help-seekers’ beliefs about how past refusals affect future compliance. In Study 1, help-seekers were more likely than potential helpers to believe that a previous refusal would lead a potential helper to deny a subsequent request of similar size. Study 2 replicated this effect, and found that help-seekers underestimated the actual compliance rate of potential helpers who had previously refused to help. Studies 3 and 4 explain this asymmetry. Whereas potential helpers’ willingness to comply with a subsequent request stems from the discomfort of rejecting others not once, but twice, help-seekers rely on dispositonal attributions of helpfulness to estimate the likelihood of hearing “yes” from someone who has previously told them “no.”