Helping you by you not helping me: Refusing help can be a ...



Helping you by not helping me: Altruistic motivations for refusing help

Daniel J. Porter & Stephanie D. Preston

University of Michigan

Author Note

Daniel J. Porter and Stephanie D. Preston, Department of Psychology, University of Michigan.

This research was supported in part by a grant from the John T. Templeton foundation and the University of Michigan to SDP. The authors thank Brian Vicker, Alicia Hofelich, and Shinobu Kitayama and the members of the Ecological Neuroscience and Culture and Cognition Laboratories for feedback on previous versions of this manuscript.

Correspondence concerning this article should be addressed to Daniel J. Porter, Department of Psychology, 530 Church Street, Ann Arbor, MI 48104.

Contact: djoepo@umich.edu

Abstract

Many studies have examined the factors that lead to helping others, yet comparatively few have investigated what causes people to reject offers of help. Most extant research in this domain has focused on selfish motivations, ignoring the potential for rejecting help for prosocial reasons, hereafter referred to as “altruistic rejection.” On the basis of spontaneous descriptions of rejection behavior and reactions to hypothetical scenarios in two studies (n = 173 & 578), we concluded that altruistic rejection exists. Individuals varied in the extent to which this motivation drives their rejection of help. People were more likely to reject aid in hypothetical situations when it was costly to the helper and they felt close to them—conditions that strengthen the interpretation of these rejections as having a prosocial quality. Perspective taking instructions allowed people to reject costly aid even when they did not feel close to the helper, again, mimicking prosocial motivations. Altruistic rejection appears to be proximately motivated by feeling distressed in the situation, which is still consistent with the empathy-altruism hypothesis. It appears that altruism influences not only when we decide to help, but also from whom we accept help, in ways we are only beginning to understand.

Keywords: altruism, help refusal, perspective taking, empathy, prosocial behavior, personal distress

.

Helping you by you not helping me: Altruistic motivations for refusing help

Evolutionarily, it makes little sense for an organism to refuse free resources, yet humans decline offers of aid every day. Although most cases are relatively inconsequential, like not taking a friend’s umbrella on a rainy day, sometimes people refuse help in situations with more dire ramifications. For instance, Ohio governor John Kasich rejected federal disaster relief after a series of destructive tornados (Horn & Faherty, 2012) and earlier a collection of African Catholic bishops refused humanitarian aid from Episcopalian sources despite widespread starvation among their parishioners (Duin, 2005). Perhaps hitting closer to home, many people know aging adults who struggle to live independently yet refuse basic help from their children or grandchildren, sometimes even failing to call for help during medical emergencies, even when they have access to phones or medical alert devices.

Though rejecting help that is needed may seem maladaptive, research has demonstrated that there really is no “free lunch,” and most help comes with a hidden price tag in the form of costs to social status, emotional well-being, or direct reciprocity obligations. In some cases, avoiding the costs of receiving help may be more important than receiving the benefits. Altruism—defined here in the biological sense of any behavior initiated by one organism that temporarily decreases its own fitness and increases or protects the fitness of another (Stephens, 1996)—appears to be costly both for helpers and recipients. Beyond the energy and resources invested by the helper, accepting aid also has a variety of explicit and implicit costs for recipients. The recipient needs to integrate these costs in order to make a sound decision to accept help, and the following formula explains a rational choice (Ackerman & Kendrick, 2008):

If ( wiBi – ( wjCj > 0, then decision = accept

Where B are benefits to the recipient, C represents costs to the recipient, and w is the magnitude of importance that an individual places on the consequence in a specific situation (Ackerman & Kendrick, 2008). Based on this type of cost-benefit analysis, it can actually be adaptive to refuse help when the costs of accepting are high. Many potential costs for accepting aid have been previously discussed in the literature including opportunity costs, energetic costs, potential injury from aggressive or deceptive conspecifics, harm to mate relationships (Ackerman & Kendrick, 2008), reduced social status (Nadler & Halabi, 2006), reduced independence (Moen, 1978), self-esteem depletion (Fisher, Nadler, & Whitcher-Algana, 1982), and obligations to reciprocate (Greenberg & Shapiro, 1971; Perugini & Galluci, 2001; Shen, Wan, & Wyer, 2011). In some cases, these problems will outweigh the benefits offered, and the rational decision is to reject the offered aid.

However, in still other cases individuals still reject help where the net benefits offered appear to outweigh the costs to that individual of accepting, and the rational decision rule appears to be violated. In non-pathological cases, individuals must perceive additional costs that are not explained by the potential costs to the recipient listed above. The only other cost that occurs in these situations happens to the helper, as he or she will usually expend resources when giving aid. We suggest that when net beneficial help is rejected, it may be because during the cost-benefit analysis, the recipient is calculating the expense the helper incurs by giving aid as a cost to his or herself. We refer to such cases where individuals refuse help, even when it would appear to be net-beneficial, as altruistic rejection—rejection that is motivated implicitly or explicitly by other-oriented desires to avoid imposing a cost on the helper, rather than self-oriented concerns about costs to the self from accepting the aid.

Though altruistic motivations have not been examined in help rejection, the ability of humans to be other-oriented and incur costs to themselves for others is well documented in other domains. For example, Hamilton’s rule (1964) demonstrates that an organism can adaptively incur a cost when it provides a greater benefit to a genetic relative, providing a kin selection mechanism for altruism to be evolutionarily stable (Smith, 1964).Extending this model to altruistic rejection, individuals may reject aid despite the lost benefits to themselves when it protects a related helper from losing important resources. Thus, grandparents may refuse aid so that their grandchildren are not “put out,”—creating a net benefit for their family—even when they may badly need that help.

Whereas the grandparents’ case more clearly falls under the rubric of Hamilton’s rule, sometimes people appear to refuse aid even when they are not related to the helpers and would not seem to incur significant costs from accepting the help. For example, when people refuse to let a good friend give them a ride to the airport and instead pay a high fee to take a taxicab, thoughts about reciprocity and independence may be relevant, but the individual may also be genuinely concerned about taking their friend’s time or energy. Although a friend is not a genetic relative, decades of research on empathy and altruism suggest that the mechanisms of prosocial giving, which evolved in the context of close, related relationships, also promote giving to non-kin, such as when we are familiar with the other, similar to them, bonded to them, or include them in our self-concept (e.g., Batson, 2011; Cialdini, Brown, Lewis, Luce, & Neuberg, 1997; Preston & de Waal, 2002; Preston, 2013). When we feel empathic towards another person, we treat them more like kin and are thus more willing to help these pseudo-kin (Ackerman, Kendrick, & Schaller, 2007). Thus, it is plausible that some of the same mechanisms that allow people to altruistically offer aid to kin and non-kin also allow altruistic rejection to be extended to kin as well as non-kin.

Altruistic rejection implies a few preconditions. First, the recipient should notice that the helper is incurring a cost, which is more likely to happen when the recipient takes the perspective of the helper. Such perspective taking (and therefore awareness of the cost to the other) is more likely to occur for close, bonded, or interdependent relationships, especially if the helper is included in the recipient’s sense of self (Galinsky, Ku, & Wang, 2005). Once the cost has been noticed, the cost to the helper also must be perceived as significant enough to outweigh the benefit to the recipient. Under these conditions, altruistic rejection is most likely.

Interestingly, being close to someone should simultaneously decrease and increase the likelihood of rejecting help via different paths. Many of the self-oriented costs that have previously been studied (e.g., physical danger, status, and reciprocity) are less relevant in close relationships (Ackerman & Kendrick, 2008). For example, while one could worry that a stranger or acquaintance offering them a ride could attack them or demand usurious repayment, this is unlikely with close friends or family. However, considerations for how helpers feel or how helping will affect them should also increase in close relationships. Taken together, people should accept more help from close others if it is low in cost or would be dangerous from a stranger, but should increasingly reject aid as the costs to the helper rise.

The current studies seek to demonstrate the existence of altruistic rejection, as well as delineate some of its proximate causes and boundary conditions. We predicted that some individuals would freely report altruistic tendencies for refusing help. Furthermore, when presented with hypothetical offers of help, we predict that offers that inconvenience the helper will be rejected more than offers that are relatively more convenient. On its own, however, simply demonstrating that people reject aid that is costly to the helper is insufficient to demonstrate the existence of altruistic rejection. It is possible that inconvenient-to-give help implies a higher debt for the recipient, so even a selfish person might be concerned with how much the helper “put out.” This may not be relevant, as Greenberg (1983) found that reciprocity obligations are scaled to the size of the benefit received and not to the cost to the other. Nonetheless, if the difference in help rejection between higher-cost and lower-cost help is altruistic, then people who are higher in altruistic traits should also altruistically reject help more. Finally, we predict that altruistic rejection should be more relevant in close relationships (kin and pseudo-kin) than in peripheral relationships. In sum, when assistance is costly to the helper, people with altruistic tendencies are expected to be more likely to reject aid if that helper is emotionally close to the recipient.

Study 1

Method

Participants. We collected 216 responses via Amazon’s Mechanical Turk (MTurk) from the United States of America. A number of responses were dropped from the study: 13 did not complete the survey, 3 responses were duplicate participants, and 27 individuals failed either trap question, indicating that they were not attending to the survey. This left 173 participants (79 males, 94 females). Age ranged from 18-75 (M = 34.53; SD = 13.01).

Procedure. Participants were recruited online from MTurk with a short description of the research and a link to the Qualtrics survey. The study was listed as “a short survey on helping behavior.” After clicking on the link, participants read an informed consent document and electronically consented to participate in the study.

To avoid participants’ answers from being influenced by suggestions of altruistic rejection, all participants completed free response blocks first and were not allowed to return to previous sections to change their answers. The study always began by asking participants to report in a paragraph a memorable instance where they had rejected help and then, in one or two words, to report their motivations for doing so. Next, participants were given a textbox to spontaneously report all easily remembered reasons for rejecting help in general. Once the participants completed the free-response section, participants rated the extent to which 14 possible reasons, generated by the researchers, influence their decisions to accept or reject help in general, (1 Not at all, 7 Very Strongly). These motivations included both self-oriented and other-oriented reasons for rejecting help (see Appendix A)

After this, participants were presented with four vignettes in random order that described offers of help from either their mother or a coworker. Each of these vignettes also had a version that was higher or lower in cost to the helper. There were four vignette scenarios: a ride to the dentist or home from the grocery store, carrying boxes, and washing dishes at a party (see Appendix B). Each vignette had four versions varying on whether their mother or a coworker offered the help, and how inconvenient it was to give. For example, in one vignette, participants were offered help carrying a large number of packages into their house. In the higher-cost version, the helper and the recipient were both wearing nice clothes and the boxes were muddy, while in the lower-cost version both people had just returned from a hike and the boxes were merely wet.

Each vignette was randomly assigned to fill one cell of a 2(Person offering: mother, coworker) x 2(Cost to helper: higher, lower) within-subjects design. So, each participant read four vignettes, a higher- and lower-cost vignette where the mother offered help, and a higher- and lower-cost vignette where a coworker offered help. Instructions at the beginning of the block asked participants to imagine themselves as best they could in these scenarios even if the specific details were not applicable to their lives. After each story, participants rated how likely they were to accept the help in the described situation on a scale from 1 (Not at all likely) to 7 (Very likely). These scores were reverse coded to create a measure of how likely they were to reject help.

To obtain participants' individual connections to their mothers and their coworkers, we had participants fill out the Inclusion of the Other in the Self scale (IOS; Aron, Aron, & Smoller, 1992). The IOS consists of sets of circles with varying degrees of overlap, and participants choose the pair of circles that best describes their relationship with an individual or group. Circles with more overlap are coded as higher values. Included among other targets, participants rated one item for their connection to their immediate family ("i.e., mothers, fathers, brothers, and sisters") and another item to someone in the same organization as them ("e.g., workplace, school, church, neighborhood"). Connection to immediate family was used to assess the connection to the mother, while connection to organizations was used for coworkers. This was included to see whether individuals who actually were more connected to the helpers (and thus more likely to perspective take) would altruistically reject help more.

We also administered some of the subscales of the abridged 30-item Penner Prosocial Personality Battery (PSB), specifically the Empathic Concern (EC), Perspective Taking (PT), and Self-Reported Altruism (SRA) subscales (Penner, 2002). The scores on these scales were summed to create a composite measure of Help Giving Motivation in order to look for connections between altruistic giving and altruistic rejection. We also included a brief demographic survey. Finally, participants were debriefed and compensated. For all analyses, α was set to .05.

Results

Participants’ free responses for a specific instance of and general reasons for rejecting help were coded in terms of containing (1) or not containing (0) an altruistic reason for rejecting help (e.g., “I don’t want to be a burden,” = 1). Two participants were not scored as they did not appear to understand the questions. 20.80% of our sample spontaneously reported altruistic motivations, and a one-sample t-test performed on these codes showed a significant difference from 0, t(170) = 6.61, p < .001.

To provide us with convergent evidence for the free-responses and with a continuous measure of the degree to which people experience prosocial motivations for refusing help, a principle components analysis (PCA) was conducted on the Survey of Help Rejection, taking all factors with eigenvalues over one. Four factors emerged, which were interpreted as Trust Concerns, Image Concerns, Altruistic Concerns, and Need Perception (see Table 1). Our factor of particular interest, Altruistic Concerns, was composed of two items, "I don't want to use up their time, money, or resources," and "I feel guilty taking their time, energy, or money."

To confirm that Altruistic Concerns scores reflected prosocial motivations, a logistic regression was used to predict the free-response codes from the four PCA factors (entered simultaneously). The two self-oriented concerns, Trust Concerns and Image Concerns, were significantly negatively predictive of free responses that were altruistic, Trust Concerns, β = -0.77, Odds Ratio (OR) = 0.46, p = .003, and Image Concerns β = -0.82, OR = 0.44, p = .001. Conversely, the Altruistic Concerns factor was significantly positively predictive of altruistic free responses, β = 1.43, OR = 4.22, p < .001. Need Perception did not significantly predict the free responses β = -0.32, p = .158. Given that the PCA factor of Altruistic Concerns continuously varies across people and predicts altruistic free responses, it was used hereafter as an individual differences measure of the tendency to perform altruistic rejection.

To further confirm the prosocial quality of altruistic rejection, we needed to determine if people are more likely to reject aid that comes at a higher than a lower cost to the helper. An omnibus test in a mixed-linear model with subjects included as a random factor found that, collapsed across all vignettes and help giver versions, high-cost help was significantly more likely to be rejected than low-cost help, F(1,680) = 96.99, p < .001. There was also a significant interaction between the cost and the vignette seen, F(3,680) = 18.26, p < .001. Because these vignettes had not been previously validated, we investigated this interaction by comparing the rates of rejection within each vignette (dentist, groceries, carrying boxes, and dish washing) collapsing across responses to mothers and coworkers. Higher-cost help was significantly more likely to be rejected than lower-cost help for a ride to the dentist, t(170) = 3.58, p < .001, a ride home from the grocery store, t(170) = 9.39, p < .001, and help carrying boxes, t(170) = 8.28, p < .001 but were not significantly different for doing dishes at a party, t(170) = 0.64, p = .525. In the dish washing vignette, both the higher (Mhigher = 4.92) and the lower (Mlower = 5.10) cost scenarios were rejected at rates similar to those for the higher-cost versions of the other three scenarios (see Figure 1). Thus, the results are interpreted to reflect an actual increase in help rejection when the aid comes at a higher cost to the helper.

To compare low versus high cost help to mothers versus coworkers within subjects, we needed all available data (from high and low cost help from all scenarios). Because we could not use the low-cost dish washing data, 40% of our sample did not have complete data for this within-subjects comparison. To retain power and still compare effects of relationship, we performed all further analyses only using data from the higher-cost scenarios, excluding cost as a variable.

To control for differences between vignettes, the amount of higher-cost help rejected was pooled across mothers and coworkers within each scenario and standardized within scenario. Then, scores were collapsed across vignettes and separate scores were created to represent the standardized level of help rejection for mothers and coworkers, which were used as the dependent variables in a repeated-measures analysis of covariance (ANCOVA) with trait Altruistic Concerns factor and trait Help Giving Motivation included as covariates. Help from mothers (M = -0.18; SD = 1.02) was rejected significantly less often than from coworkers (M = .19; SD = 0.93), main effect of likelihood of rejecting help: F(1,169) = 13.06, p < .001. However, trait Altruistic Concerns predicted rejecting higher-cost help from mothers, B = 0.21, t(170) = 2.70, p = .008, but not from coworkers, B = 0.05, t(170) = 0.64, p = .524. Help Giving Motivation did correlate with Altruistic Concerns, r(172) = .19, p = .010, though it did not predict help rejection from mothers, B = 0.003, t(170) = 0.30, p = .764, or from coworkers, B = 0.003, t(170) = 0.34, p = .736.

Our a priori reason for including mothers and coworkers as the helpers was that they naturally vary in the extent to which people feel connected to them, and this difference may explain why Altruistic Concerns was differentially predictive between them. To confirm this prediction, a mixed-linear model with subject included as a random effect was conducted looking at the interaction between the IOS and Altruistic Concerns, collapsed across mothers and coworkers. As expected, the omnibus test was found to be significant, where Altruistic Concerns seemed to predict help rejection when the helper was highly connected, but not weakly connected to the recipient, FIOS*AltruisticConcerns(1,340) = 9.06, p = .003. To show that this held for all groups (e.g., weakly connected mothers and highly connected coworkers), we created categories from the specific IOS responses for participants' connection to organizations (M = 2.83; SD = 1.40) and immediate family (M = 5.08; SD = 1.78). IOS responses one standard deviation or more above the mean for that group were coded as high connection, and responses one standard deviation or more below the mean were coded as low connection. Due to unequal sample sizes, regressions were run within each of the four groups (high or low connection x mothers or coworkers), with the help rejected from that helper as the dependent variable and Altruistic Concerns as the predictor. When the participant felt highly connected to the helper, Altruistic Concerns predicted rejecting higher-cost help, βMotherHigh = 0.25, p = .010, βCoworkerHigh = 0.47, p = .040, but when they felt less connected to the helper, altruistic rejection did not predict the amount of help rejected, βMotherLow = 0.05, p = .800, βCoworkerLow = 0.08, p = .461 (see Figure 2).

Discussion

The purpose of the current study was to demonstrate for the first time the existence of altruistic rejection. Through their open-ended free responses, a significant portion of participants spontaneously indicated that concern for the helper is salient when deciding to accept or reject help. Convergent evidence from a questionnaire about different possible reasons for rejecting help also revealed a stable factor of Altruistic Concerns. Together, this demonstrates that the cost to the helper matters to some individuals. Moreover, in three of four hypothetical vignettes, people were more likely to reject the aid that was more costly to the helper—a relationship that is likely to occur only when the recipient of the aid is considering how the help will affect the helper. By definition, this is an “other-oriented” or prosocial concern. Further reflecting effects that occur in typical experiments on altruistic giving, participants’ trait Altruistic Concerns predicted rejecting higher cost aid only when they felt more connected to the helper, regardless of whether it was their mother or coworker, which was supported by ratings of closeness to relatives and people in their organizations on the IOS. As further evidence that altruistic rejection is a real phenomenon, the Altruistic Concerns factor was significantly correlated with trait measures of empathy and altruism. However, trait empathy and altruism did not predict help rejection, suggesting that the tendency to reject aid based on perceived costs to the other is a related but unique construct that has its own explanatory power for when we reject help. Study 2 was designed to examine the possibility that altruistic rejection, unlike traditional empathy and altruism, is promoted in situations of distress rather than empathic concern per se.

One interesting result was the fact that people actually accepted more help from their mothers when it was higher cost. As mentioned previously, aid from close others should be accepted more often when it is low in cost or beneficial to their shared outcomes because many of the costs of receiving aid are negligible in these cases (e.g., fear of being taken advantage of or viewed as weak). Furthermore, even the higher-cost situations used in this study were fairly mundane and low in cost compared to life-saving or truly consequential types of help. Thus, the costs to the helper were perhaps too low in this case to outweigh the reduced costs of accepting aid from a close other, particularly a mother, from whom one may be accustomed to receiving help. It may be the case that at particularly high costs of helping, it is culturally appropriate to accept help from relatives only, and so for the current set of studies we continued using these low-cost scenarios.

Because our participants changed their likelihood of rejecting aid across cost conditions and relationships, it is unlikely that their behavior reflects a more basic tendency to reject all aid, or a social desirability bias, but Study 2 additionally measured social desirability to confirm this. In addition, we assume that people reject aid more in close relationships because they are more likely to take the perspective of the helper and perceive the costs to them in this case, and so Study 2 included a perspective-taking manipulation to further confirm this assumption.

Study 2

In Study 2, we hoped to address several unanswered questions from Study 1. First, due to the failed cost manipulation in one of our vignettes, we were unable to show that our Altruistic Concerns factor only predicted rejecting high-cost help and was not simply a tendency to reject help regardless of how it affects the helper.

Additionally, in order to altruistically reject help, the recipient needs to notice a cost to the helper, which likely requires the helper to take the perspective of the recipient. Once they notice this cost, they also need to care that the helper will lose resources. In Study 1, we found some evidence of this where a trait Altruistic Concerns factor predicted rejecting more help from close, but not distant others. Usually perspective taking is more common in close relationships, however, this does not directly address the issue of perspective taking. For our second study, we chose to directly induce perspective taking by modifying Batson et al.’s (1997) procedure. Their procedure uses two different instructions, asking participants to either imagine how the other person is feeling, or to imagine how they would feel in the other person’s shoes.

We also do not know what proximate emotions motivate altruistic rejection. In the literature on altruistic giving, there has been much work on the emotions of empathic concern (EC) and personal distress (PD) (Batson, O’Quin, Fultz, Vanderplas, & Isen, 1983; Baston, Early, & Salvarani, 1997; Eisenberg, Fabes, Schaller, & Miller, 1987; Preston, Hofelich, & Stansfield, 2013). Empathic concern is more other-oriented, and includes warm, tenderhearted, compassionate feelings, while personal distress is more self-focused sensations of being upset, worried, or troubled (Batson et al., 1997). While both of these emotions have been found to motivate altruistic giving, empathic concern is usually a better predictor (Batson et al., 1997; Preston et al., 2013), and individuals feeling personal distress will take the opportunity to escape rather than help if given the chance (Batson et al., 1983). In the case of altruistic rejection, although the first intuitive prediction might be that empathic concern will also better predict refusing help, there are reasons to believe that personal distress may play the dominant role.

Fundamentally, giving help requires that two individuals interact socially. On the other hand, help rejection at its core is the denial of a social interaction. Given that distress encourages escape when the option is present, whereas empathic concern leads to giving help even when allowed to get away, empathic concern and personal distress may additionally vary on the dimension of social approach versus avoidance. In a giving context, the altruistic response requires approaching the other person, and so it is logical that empathy will be a better predictor for altruistic help giving. However, as the recipient, the altruistic behavior is to reject the help and deny the social interaction. In this case, an approach motivation (empathic concern) will actually decrease the likelihood of behaving altruistically, while an avoidance motivation (personal distress) would encourage the altruistic response. Thus, we predict that emotions of personal distress will predict help rejection better than empathic concern.

We have several predictions for Study 2. First, as we found in three out of four vignettes in Study 1, aid that inconveniences the helper (higher-cost help) should be rejected more than relatively convenient (lower-cost) help. Secondly, trait variables related to altruistic rejection (e.g., Altruistic Concerns) should be important in higher-cost, but not lower-cost help offers. Because we predict that altruistic rejection will only be seen in higher-cost situations, the interaction we observed in Study 1, where trait Altruistic Concerns predicted help rejection only for close others, would also only be seen in the higher-cost offers. Furthermore, because we believe that the reason Altruistic Concerns only predicted help rejection for close others was because the participants were taking their perspectives, we hypothesized that this interaction would not be seen when perspective taking was induced. Regarding EC and PD, we predicted that personal distress would relate to help rejection more than empathic concern.

Method

Participants. We collected 648 responses from Amazon’s Mechanical Turk. Duplicate submissions were found for 25 responses, and another 44 failed trap questions or self reported paying little attention to the survey, leaving 579 participants (277 male, 301 female, 1 other). Age ranged from 18-75 years (M = 34.33; SD = 12.52).

Materials. The Penner Prosocial Battery and Survey of Help Rejection scales from the previous experiment were used again in this study. Additionally, participants only saw the two best vignettes from Study 1 (ride home from groceries and carrying boxes), but in this case offers of help always came from a friend. No specific instructions were given regarding what kind of friend the participant should have imagined. For Study 2, we included the X-1 shortform of the Social Desirability Scale (SDS; Crowne & Marlowe, 1960) and the Emotional Response Questionnaire, which asks participants to rate the extent to which they are currently feeling 26 emotional adjectives (ERQ; Batson et al., 1997). Participants who completed the survey after the first wave of data collection also answered an item for each cost level assessing ‘how much they felt the person offering help in this scenario would be “put out” by helping you.’

Procedure. Participants first filled out the PSB, SDS, and the two vignettes in random order. After this, they filled out the Survey of Help Rejection and the demographics. Participants after the initial round of data collection now also answered the question assessing their perception of the cost to the helper for each scenario they saw.

The vignette sections differed from Study 1 in several ways. All offers of help came from a friend rather than from their mothers or coworkers, and for each vignette participants filled out the ERQ and rated their connection on the IOS to the specific person they imagined offering them help. Most importantly, participants saw one of three instructions for reading the vignettes. One third of participants received the same instructions from Study 1, which asked them to “imagine how you would feel if you were in this situation.” We called this replication the Imagine Own Feelings condition. The rest of the participants received one of two perspective taking manipulations which have previously been used in examining altruistic giving (Batson et al., 1997). In one, participants were instructed to “imagine how the person offering you help feels in this situation,” (Imagine Other’s Feelings), and in the other participants were told to “imagine how you yourself would feel if you were the person offering help,” (Imagine Other’s Shoes).

Results

Replicating Study 1. To confirm the validity of our results from Study 1, we reran all replicable tests in Study 2. First, we conducted a PCA on the new participants’ responses to the Survey of Help Rejection. The Image Concerns, Trust Concerns, and Altruistic Concerns factors re-emerged with minor variation in the items and weights (see Table 2)[1]. As in Study 1, when higher-cost help was offered, and participants were not given perspective taking instructions, the interaction between Altruistic Concerns and the IOS was significant, F(1,177) = 21.89, p < .001. When participants were highly-connected to the helper (IOS > 1 SD from mean), Altruistic Concerns predicted rejecting more help, β = 0.57, t(30) = 3.84, p = .001, but when they were not very connected (IOS < 1 SD from mean), it did not predict help rejection, β = 0.00, t(31) = -0.01, p = .993.

ERQ Factors. Based on previous work that has shown more than two factors on the ERQ (Preston et al., 2013), we performed a factor analysis on responses to assess the underlying dimensions in our sample. Four factors emerged from this analysis and differ from Batson et al. (1997), but replicate other research (Preston et al., 2013). We interpreted these factors as Distressed, Pleased, Horrified, and Compassionate (see Table 3). Most of the PD items loaded onto Distressed (along with a few extras), while the EC items were split between the Pleased and Compassionate factors.

Cost of Helping. In Study 2, we wished to answer the questions related to the cost to the helper, which we were unable to assess in Study 1. As a manipulation check, a paired samples t-test showed that, as predicted, higher-cost help was rejected significantly more than lower-cost help (Mhigher = 2.80; Mlower = 1.62; SDhigher = 1.71 SDlower = 1.04), t(578) = 16.27, p < .001.

Survey of Help Rejection Factors & the IOS. We collapsed across perspective taking instructions and ran regressions on the higher- and lower-cost help offers using the three trait Survey of Help Rejection factors (Image Concerns, Trust Concerns, and Altruistic Concerns), the IOS, and our four state ERQ factors as predictors. Due to our prediction that the state ERQ variables might mediate the trait variables, these predictors were entered in two steps. The traits were included alone first, and then we added the ERQ factors in a second step.. Because we were primarily interested in which set of factors significantly predicted help rejection in the cost conditions, rather than magnitude differences between the same factors, we conducted these as two separate regressions for higher- and lower-cost help.

As we hypothesized, Altruistic Concerns significantly predicted rejecting higher-cost offers, β = 0.23, t(565) = 5.84, p < .001, but did not significantly predict rejecting lower-cost help, β = 0.04, t(565) = 0.96, p = .335. Also as we found in Study 1, Image Concerns did not predict the amount of higher-cost help rejected, β = 0.02, t(565) = 0.41 p = .680, however, we found that Image Concerns did predict rejecting lower-cost help, β = 0.19, t(565) = 4.90, p < .001. Trust Concerns related to neither type of help offered, βhigher = 0.04, t(565) = 1.06, p = .290, βlower = 0.02, t(565) = 0.42, p = .674. The IOS predicted less rejection of help for both higher-cost offers, β = -0.23, t(565) = 5.73, p < .001, and lower-cost offers, β = -0.19, t(565) = -4.73, p < .001.

When the state ERQ factors were added to the model, the results in the higher-cost analysis remained stable, but two predictors in the lower-cost analysis changed their level of significance, p < .10, p < .05, p < .01, p < .001. Altruistic Concerns p value came much closer to significance than before, βstep-two = 0.07, t(565) = 1.78, p = .076. Image Concerns still significantly predicted rejecting lower cost offers but with a smaller coefficient, β = 0.10, t(565) = 2.54, p = .011.

Emotion Related Questionnaire Factors. Feeling Distressed predicted rejecting higher-cost, β = 0.10, t(565) = 2.50, p = .013, but not lower-cost help, β = .04, t(565) = 0.60, p = .362. Being Pleased predicted less rejection of higher-cost, β = -0.23, t(565) = -5.72, p < .001, and lower-cost help, β = -0.24, t(565) = -5.96, p < .001. Feeling Horrified predicted rejecting higher-cost help, β = 0.09, t(565) = 2.35, p = .019, and lower-cost help, β = 0.30, t(565) = 7.95, p < .001. Compassion was unrelated to rejecting any help (see Table 4).

Social Desirability. To rule out the possibility that altruistic rejection was simply due to the effect of social desirability, we added our measure of this trait to the model. The effects reported above did not change significantly, and social desirability did not predict rejecting higher-cost help, β = 0.03, t(564) = 0.82, p = .414, and marginally predicted rejecting lower-cost help, β = 0.07, t(564) = 1.84, p = .066.

Subjective Perception of Cost. As a manipulation check, a paired-samples t-test showed that higher-cost help (M = 4.30; SD = 1.48) was seen as significantly more costly to the helper than lower-cost help (M = 2.97; SD = 1.47), t(357) = 14.40, p < .001. To test the hypothesis that perceiving a cost to the helper can lead to rejecting help, we added the direct measure of cost perception to the regression model. Due to having fewer responses on this item than for others questions, this regression was run independently from the other analyses. Participants subjective perception of how costly the offer was to the helper predicted rejecting both higher-cost, β = 0.20, t(348) = 3.94, p < .001, and lower-cost help, β = 0.14, t(348) = 2.74, p = .006.

Mediating Effects. As predicted, personal distress (our Distressed factor), but not empathic concern (split between our Pleased and Compassionate factors), predicted rejecting higher-cost help. To provide further evidence that these feelings were motivating altruistic, not selfish behavior, we conducted a Sobel test with Distressed feelings mediating the relationship between trait Altruistic Concerns and higher-cost help rejected. We used Andrew Hayes’ SPSS SOBEL macro to perform these tests (Preacher & Hayes, 2003). The direct effect of Altruistic Concerns on higher-cost help accepted was βdirect = 0.37, p < .001, which was significantly partially mediated by the indirect effect of Distressed, βindirect = 0.03, Z = 2.13, p = .033 (see Figure 3). Feelings of Horror also predicted rejecting higher-cost help, but these did not mediate the relationship between Altruistic Concerns and higher-cost help rejected, βindirect = 0.01, Z = 1.44, p = .155.

We also found that the self-oriented trait Image Concerns predicted increased rejection of lower-cost help. We performed the same mediations as above using Image Concerns and lower cost help. The direct effect of Image Concerns on lower-cost help was, βdirect = 0.22, p < .001. Distress did not mediate this relationship, β = 0.00, Z = -0.19, p = .849. However, feeling Horrified did significantly mediate Image Concerns, βindirect = 0.07, Z = 4.61, p < .001 (see Figure 4).

We also used participant’s subjective perception of cost as a mediator between the help rejected and the relevant SHR trait associated with the level of cost (Altruistic Concerns for higher-cost, Image Concerns for lower cost). The perception of cost marginally mediated the effect of Altruistic Concerns on help rejection, βindirect = .04, Z = 1.90, p = .057. The perception of cost in lower-cost offers significantly mediated Image Concerns, β = .03, Z = 2.13, p = .033.

Perspective Taking. We hypothesized that the interaction between the IOS and trait Altruistic Concerns found in Study 1 was due the tendency to take the perspective of highly-connected, but not weakly-connected others. To test this hypothesis, we ran a mixed linear model with a four-way interaction between inconvenience, perspective taking condition, IOS, and trait Altruistic Concerns included as a term. Subject was included as a random effect. This four-way interaction was significant, F(6,1145) = 6.54, p < .001, and as predicted the interaction between Altruistic Concerns and the IOS from Study 1, where Altruistic Concerns predicted rejecting assistance only when the helper was included in the recipient’s sense of self, was only seen when inconvenience was high and perspective taking was not induced. When perspective taking was induced, Altruistic Concerns predicted rejecting higher-cost aid from helpers who were both high and low on the IOS (see Figure 5).

Reciprocity. Although a reciprocity related factor did not emerge from our PCA, an item "I don't want to have to return the favor" was included in the Survey of Help Rejection. Looking at this item individually, it did not correlate with the likelihood of higher-cost help rejected, r(579) = .05, p = .204, but it did significantly correlate with the likelihood of rejecting lower-cost help, r(579) = .16, p < .001.

Discussion

In Study 2, we were able to replicate and extend our findings from Study 1. We once again found that a trait of Altruistic Concerns emerged from a PCA of the Survey of Help Rejection. In Study 1, Altruistic Concerns predicted rejecting higher-cost help, but only when the recipient was highly connected to the helper. We hypothesized that this was due to the tendency to perspective take and notice the cost with close-others. In Study 2, when perspective taking was not induced, we replicated this interaction. However, as predicted, when perspective taking was induced, Altruistic Concerns predicted rejecting help from both highly and weakly connected helpers. We took this as evidence that the reason Altruistic Concerns differentially predicted rejecting help for close and non-close helpers in Study 1 was because people are more likely to spontaneously take the perspective of someone included in their sense of self.

We found a double dissociation where Altruistic Concerns and feeling Distressed predicted rejecting offers of aid that were relatively higher-cost to the helper, but not offers which were lower-cost, and feeling Distressed mediated Altruistic Concerns relationship with the help rejected here. Image Concerns and feeling Horrified showed the opposite pattern, where they predicted rejecting help when it was relatively lower-cost, but not when it was higher-cost, and feeling Horrified mediated the relationship between Image Concerns and help rejected.

In support of our hypothesis, we found that empathic concern (split between our Pleased and Compassionate factors) did not predict help rejection, while feeling distressed did. This converges with our results and conclusions from Study 1, where a trait prosociality measure, a significant component of which is trait empathic concern, did not predict help rejection. We believed that empathic concern was a strong predictor of altruistic giving because it encouraged social approach. However, in help rejection, the altruistic action requires social avoidance, promoted by personal distress, and empathic concern might actually lead one to accept the help. This was supported by our findings. Personal distress predicted help rejection, and partially mediated the relationship between help rejection and an altruistic trait (Altruistic Concerns). On the other hand, part of the construct of empathic concern, the Pleased factor, actually predicted accepting more help and therefore imposing a cost on the helper. This warm, affiliative response may reflect a reinforcing effect on social bonds which receiving an offer of help from a close other likely provides.

We found that people do use information about the cost to others when deciding to reject help. It is possible that although this appears to be altruistic, it could still be motivated by selfish concerns. If the obligation to reciprocate increases with the cost to the helper, then even a selfish person would be concerned with inconveniencing their benefactor. However, Greenberg (1980) showed that it is the benefit received, not the effort expended by the helper, which determines feelings of reciprocity. Furthermore, in examning the item from the Survey of Help Rejection related to reciprocity concerns, we found it did not predict higher-cost help, but it did predict rejecting lower-cost help. So, while reciprocity is certainly a factor in decisions to refuse help, it does not explain our phenomenon.

Finally, social desirability did not predict help rejection when the aid was costly to the helper, and only had a marginal effect when it was less costly. Given this, as well as the inability for social desirability to explain the pattern of results in Study 1, we do not believe that it is the driving force behind altruistic rejection.

General Discussion

The present work provides preliminary evidence for the hereto neglected phenomenon that individuals can be other-oriented when rejecting help. In Study 1, some individuals freely reported thinking about the impact that accepting aid would have on the helper. In both experiments, a factor that indicated altruistic concern for the helper emerged from a principle components analysis of motivations generated by our lab. Taken as a trait variable, this factor predicted rejecting assistance that would inconvenience the helper, but only if the helper was included in the recipient’s sense of self. By inducing perspective taking however, we were able to extend this help rejection to non-close others.

Proximately, we found feeling distressed predicted rejecting help, and it also accounted for part of the relationship between the altruistic factor from the PCA and help rejection. This goes against the traditional story in altruistic giving, where feeling empathic concern for a target predicts prosocial behavior better than personal distress (Batson et al., 1997). However, being altruistic in the domains of giving and receiving requires very different behaviors, namely engaging in social interaction to altruistically give help versus disengaging from a social interaction to altruistically reject help. Since the construct of empathic concern includes words related to feeling close to others and social approach motivations, whereas personal distress has been shown to promote social avoidance in giving situations, we believe that these results do not in fact contradict the predictions of the empathy-altruism hypothesis.

However, other researchers have made a distinction between emotions felt for oneself and for others in relation to altruism (Batson et al., 1997). Though distress does seem to be motivating altruistic behavior, it would be more convincing to call this a prosocial feeling if we could demonstrate that our participants are feeling distressed for the helper, rather than just feeling distressed themselves (Batson et al., 1997). Though we cannot address this in our current data, we plan to examine this in a follow-up study.

We believe from our current evidence that altruistic rejection is very likely to exist, however there are several limitations with our studies that must be addressed. First, all of our data was collected online, allowing us to obtain a diverse sample but limiting our control over the environment in which participants completed our survey. Furthermore, our measures of help rejection relied on self-reported reactions to hypothetical scenarios, and could have been influenced by social desirability. However, both the pattern of results and the direct measurement of this was inconsistent with this explanation in both studies. These types of hypothetical or imaginative situations are commonly used to study altruism (e.g., Batson et al., 1997; Preston et al., 2013), but we are also currently conducting a study to test subjects in a controlled laboratory using real offers of help.

From the perspective of society, whether altruistic rejection is actually a desirable behavior likely depends greatly on context. If the behavior could be promoted broadly in community settings, it might help limit consumption of scarce resources, an issue that will only continue to become more important as time progresses. However, for most people personally, offers of help are probably made sincerely to our friends and family, and we do not want to see that person remain in need. In this case, altruistic rejection is probably not something we would like to encourage. In some cases, such as with older adults refusing much needed aid, we may even wish to intervene. Given this, geriatric care practices may benefit from re-evaluation. For example, it may be advisable for children and grandchildren to present offers of help in ways that minimize the costs to themselves, while non-kin workers might be better served intervening at the previously studied reasons for refusing help, such as emphasizing the patient’s autonomy.

We believe the current research presents evidence for the existence of prosocial feelings when choosing to reject help and demonstrates another domain in which humans can act for the benefit of others. Moving forward, we hope to further explore the proximate causes and delineate the boundary conditions of this phenomenon. By better understanding these phenomena, we can attempt to maximize our ability to effectively care for those in need.

References

Ackerman, J. M., & Kendrick, D. T. (2008) The costs of benefits: Help-refusals highlight key trade-offs of social life. Personality & Social Psychology Review, 12, 118-141.

Ackerman, J. M., & Kendrick, D. T., & Schaller, M. (2007). Is friendship akin to kinship? Evolution & Human Behavior, 28, 365-364.

Aron, A., Aron E. N., & Smollan, D. (1992). Inclusion of other in the self scale and the structure of interpersonal closeness. Journal of Personality and Social Psychology, 63, 596-612.

Batson, C. D. (2011). Altruism in humans. USA: Oxford University Press.

Batson, D. C., Early, S., & Salvarani, G. (1997). Perspective taking: Imagining how another feels versus imagining how you would feel. Personality and Social Psychology Bulletin, 23, 751-758.

Batson, C. D., O'Quin, K., Fultz, J., Vanderplas, M., & Isen, A. M. (1983). Influence of self-reported distress and empathy empathy on eqoistic versus altruistic motivation to help. Journal of Personality and Social Psychology, 45, 706-718.

Cialdini, R. B., Brown, S. L., Lewis, B. P., Luce, C., & Neuberg, S. L. (1997). Reinterpreting the empathy–altruism relationship: When one into one equals oneness. Journal of Personality & Social Psychology, 73, 481-494. doi:10.1037/0022-3514.73.3.481

Clark, M. S., & Mills, J. R. (2011). A theory of communal (and exchange) relationships. In P. Van Lange, A. Kruglanski, & T. Higgins (Eds.), Handbook of Theories of Social Psychology. Los Angeles, CA: Sage.

Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24(4), 349-354.

Duin, J. (2005, June 7). African bishops reject aid. The Washington Times. Retrieved from

Eisenberg, N., Fabes, R. A., Schaller, M., & Miller, P. A. (1989). Sympathy and personal distress: Development, gender differences, and interrelations of indexes. New Directions For Child Development, 44, 107-126. doi:10.1002/cd.23219894408

Fisher, J. D., & Nadler, A. (1976) Effect of donor resource on recipient self-esteem and self help. Journal of Experimental Social Psychology, 12, 139-150.

Fisher, J. D., Nadler, A., & Whitcher-Algana, S. (1982) Recipient reactions to aid. Psychological Bulletin, 91, 27-54.

Galinsky, A. D., Ku, G., & Wang, C. S. (2005). Perspective-Taking and Self-Other Overlap: Fostering Social Bonds and Facilitating Social Coordination. Group Processes & Intergroup Relations, 8), 109-124. doi:10.1177/1368430205051060

Greenberg, M. S., & Shapiro, S. P. (1971). Indebtedness: An adverse aspect of asking for and receiving help. Sociometry, 34, 290-301.

Greenberg, M. S., & Silverman, M. A. (1971). Determinants of helping behavior: Person's rewards versus Other's costs. Journal of Personality, 39, 79-93.

Hamilton, W. D. (1964). The genetical evolution of social behavior. Journal of Theoretical Biology, 7, 1-52.

Horn, D., & Faherty, J. (2012, March 3). Kasich turns down federal disaster aid. . Retrieved from

Mathur, V. A., Harada, T., Lipke, T., & Chiao, J. Y. (2010). Neural basis of extraordinary empathy and altruistic motivation. NeuroImage, 51, 1468-1475.

Moen, E. (1978) The reluctance of the elderly to accept help. Social Problems, 25, 293-303.

Nadler, A., & Halabi, S. (2006). Intergroup helping as status relations: Effects of status stability, identification, and type of help on receptivity to high status group’s help. Journal of Personality & Social Psychology, 91, 97-110.

Penner, L. A. (2002). The causes of sustained volunteerism: An interactionist perspective. Journal of Social Issues, 58, 447-468.

Perugini, M., & Gallucci, M. (2001). Individual differences and social norms: The distinction between reciprocators and prosocials. European Journal of Personality, 15, S19-S35.

Preston, S. D., & de Waal, F. B. M. (2002). Empathy: Its ultimate and proximate bases. Behavior and Brain Sciences, 25, 1-72.

Preston, S. D., Hofelich, A. J., & Stansfield, R. B. (2013). The ethology of empathy: A taxonomy of real-world targets of need and their effect on observers.

Shen, H., Wan, F., & Wyer, R. S. Jr. (2011). Cross-cultural differences in refusal to accept a small gift: The differential influence of reciprocity norms on Asians and North Americans. Journal of Personality and Social Psychology, 100, 271-281.

Stewart-Williams, S. (2007). Altruism among kin vs. non-kin: Effects of cost of help and reciprocal exchange. Evolution and Human Behavior, 28, 193-198.

Singelis, T. M., Triandis, H. C , Bhawuk, D., & Gelfand, M. J. (1995). Horizontal and vertical dimensions of individualism and collectivism: A theoretical and measurement refinement. Cross-Cultural Research, 29, 240-275.

Stephens, C. (1996). Modeling reciprocal altruism. British Journal for the Philosophy of Science, 47, 533–551

Triandis, H. C., & Gelfand, M. J. (1998). Converging measurement of horizontal and vertical individualism and collectivism. Journal of Personality and Social Psychology, 74, 118-128.

Appendix A

For each reason listed below, please indicate how much it influences your decision to accept or reject help.

1. It feels offensive

2. It implies that they are superior

3. I don’t think I need help

4. I don’t want to have to return the favor,

5. I don’t want to use up the person’s time, money, or resources

6. It implies I am not independent

7. I feel embarrassed or singled out when someone offers me help

8. I don’t trust their intentions

9. I don’t like them, or want to get involved with them

10. I feel guilty taking their time, energy, or money

11. I don’t want to be dependent on others

12. I have never helped them before/we don’t have that type of relationship

13. They seem like they were only offering to be polite/their offer didn’t seem sincere

14. It would make our relationship unbalanced.

Appendix B

Vignette Scenarios - Text was the same for mother and coworker offers except for the noun used.

Ride to Dentist:

Higher-cost: You are chatting with your [mother/coworker] about plans for tomorrow. She will be spending her day working on a report that her boss wants by the end of the week. She may need to stay late that evening to get enough done on the project. You describe your dental appointment coming up tomorrow afternoon, and explain that you have to take the bus because your car is being fixed. She remarks that the bus is unreliable and offers to leave work for a bit that afternoon to drive you to your appointment.

Lower-cost: You are chatting with your [mother/coworker] about plans for tomorrow. She will be spending her day working on a report that her boss wants by the end of the month. She may also stay late that evening to search for flights for a vacation next summer. You describe your dental appointment coming up just before work the next morning, and explain that you have to take the bus because your car is being fixed. She remarks that the bus is unreliable and offers to come by your place on the way to work to drive you to your appointment.

Ride Home from Grocery Store:

Higher-cost: You have just gone grocery shopping, and as you were paying, the bus came and left early. You aren’t sure when the next bus will come, so you are just waiting at the stop in front of the store. A vehicle approaches and rolls down its window—it’s your [mother/coworker]. She says that she is on her way to her exercise class. She offers to drop you off at your house and then head back to her class, missing the first 10 or 15 minutes of the hour-long session.

Lower-cost: You have just gone grocery shopping, and as you were paying, the bus came and left early. You aren’t sure when the next bus will come, so you are just waiting at the stop in front of the store. A vehicle approaches and rolls down its window—it’s your [mother/coworker]. She says that she is returning from her exercise class and doing some errands. She offers to drop you off at your house on the way to her first stop, probably the paint store.

Carrying Boxes:

Higher-cost: You have just gone for a nice lunch with your [mother/coworker], and you are both dressed in nice outfits for the occasion. Walking back towards your house, you see that a pile of many small boxes was delivered to your doorstep. Unfortunately, the delivery man left them in the mulch next to the steps where the sprinkler was running. You pick one up and realize that boxes are muddy, and you will probably get dirty bringing them inside. Seeing that there are multiple trips worth of packages, your [mother/coworker] offers to help you carry them in.

Lower-cost: You have just gone for a light hike with your [mother/coworker], and you are both dressed in old clothes and boots. Walking back towards your house, you see that a pile of many small boxes was delivered to your doorstep. Unfortunately, the delivery man left them on the yard next to the steps where it had rained. You pick one up and realize that the boxes are a little wet, but the contents should be safe. Seeing that there are multiple trips worth of packages, your [mother/coworker]offers to help you carry them in.

Washing Dishes at a Party:

Higher-cost: You are having an early evening party for your friends, family, and coworkers. Everyone is enjoying themselves, so you decide to clean up a few dishes in the kitchen. Your [mother/coworker] walks in looking for the trash, and you strike up a conversation. She is planning to leave in a few minutes to meet some friends. But then she notices the dishes and offers to stay at the party a little longer to help you clean them.

Lower-cost: You are having an early evening party for your friends, family, and coworkers. Everyone is enjoying themselves, so you decide to clean up a few dishes in the kitchen. Your [mother/coworker] walks in looking for the trash, and you strike up a conversation. She is planning to meet some friends later that night. She notices you doing the dishes and offers to stay in the kitchen a little longer to help you clean them.

Table 1

Principle Components Analysis of Survey of Help Rejection in Study 1

|Factor 1: Mistrust | |Factor 2: Image Concerns | |

|(29%) | |(16%) | |

|I don't trust their intentions. |.799 |It implies that I am not independent. |.765 |

|I don't like them, or don't want to get involved with |.769 |I don't want to be dependent on others. |.691 |

|them. | | | |

|They seemed like they were only offering to be |.730 |I feel embarrassed or singled out when someone offers |.688 |

|polite/their offer didn't seem sincere. | |me help. | |

|I never helped them before/We don't have that type of |.638 |It implies that they are superior. |.651 |

|relationship. | | | |

| | | | |

|Factor 3: Altruistic Rejection (10%) | |Factor 4: Need Perception | |

| | |(7%) | |

|I feel guilty taking their time, energy, or money. |.890 |I don't think I need help. |.918 |

|I don't want to use up the person's time, money, or |.875 | | |

|resources. | | | |

Note. Percentage of variance explained is listed, as well as the top four rotated coefficients with absolute value > .5. The SHR is composed of 14 total items.

Table 2

Principle Components Analysis of Survey of Help Rejection in Study 1

|Factor 1: Image Concerns | |Factor 2: Mistrust | |

|(41%) | |(12%) | |

|It implies that they are superior. |.800 |I don't like them, or don't want to get involved with |.837 |

| | |them. | |

|It implies that I am not capable. |.742 |They seemed like they were only offering to be |.742 |

| | |polite/their offer didn't seem sincere. | |

|It would make our relationship unbalanced. |.700 |I don't trust their intentions. |.734 |

|I feel embarrassed or singled out when someone offers me |.579 |I never helped them before/We don't have that type of |.508 |

|help. | |relationship. | |

| | | | |

|Factor 3: Altruistic Rejection | | | |

|(8%) | | | |

|I don't want to use up the person's time, money, or |.865 | | |

|resources. | | | |

|I feel guilty taking their time, energy, or money. |.862 | | |

|I don't want to be dependent on others. |.605 | | |

|I feel embarrassed or singled out when someone offers me |.504 | | |

|help. | | | |

Note. Percentage of variance explained is listed, as well as the top four rotated components with absolute value > .5. The SHR is composed of 14 total items.

The item “It implies I am not independent” in Study 1 was changed to “It implies that I am not capable” for Study 2.

Table 3

Principle Components Analysis of the Emotional Response Questionnaire

|Factor 1: Distressed | |Factor 2: Pleased | |

|(39%) | |(17%) | |

|Upset |.864 |Pleased |.847 |

|Distressed |.834 |Happy |.821 |

|Worried |.817 |Warm |.784 |

|Bothered |.814 |Moved |.781 |

|Troubled |.771 |Tender |.557 |

|Concerned |.761 |Softhearted |.520 |

| | | | |

|Factor 3: Horrified |8% |Factor 4: Compassionate | |

|(5%) | |(4%) | |

|Horrified |.764 |Sympathetic |.826 |

|Sorrowful |.721 |Compassionate |.760 |

|Grieved |.596 |Softhearted |.670 |

|Afraid |.585 |Tender |.559 |

|Sad |.538 | | |

|Panicked |.523 | | |

Note. Percentage of variance explained is listed, as well as the top six rotated components with absolute value > .5. The ERQ is composed of 26 total items.

Table 4.

SHR and ERQ Factors Which Predict Rejecting Higher-cost and Lower-cost Help Offers

Higher-Cost Lower-Cost

Image Concerns -0.04 .19*** (0.10*)

Trust Concerns 0.01 0.00

Altruistic Concerns 0.23*** .04 (0.07)

IOS -0.23*** -0.19***

______________________________________________________________________

Distressed 0.10* 0.02

Pleased -0.24*** -0.23***

Horrified 0.10* 0.28***

Compassionate 0.00 0.03

Cost Perception[2] 0.20*** 0.14**

Note. Significance levels * p < .05 ** p < .01 *** p < .001

β values predicting help rejected for higher- and lower-cost offers

Factors above the line were entered in the first step of the regression; values in parentheses indicate p values which changed their significance level in the second step

[pic]

Figure 1. Offers which were higher-cost to the helper were rejected significantly more often than lower-cost offers in three out of four scenarios. The lower cost-version of the dishes scenario seems to have been seen as also higher-cost, rather than the reverse.

[pic]

Figure 2. Altruistic rejection predicts accepting less help when the participant is highly connected to the person giving the help. Connection to Helper categories were created from IOS response for each helper, where a Z-score > +1 is a high connection and < -1 is a low connection.

[pic]

Figure 3. Feeling Distressed significantly partially mediates the trait Altruistic Concerns ability to predict increased help rejection of higher-cost help.

[pic]

Figure 4. Feeling Horrified significantly partially mediates the trait Image Concerns ability to predict rejecting lower-cost help.

[pic]

Figure 5. When asked to imagine the scenarios from their own perspective (Study 1 replication), Altruistic Concerns predicted rejecting help from highly connected, but not weakly connected helpers. When perspective taking was induced, this interaction disappeared and the trait predicted rejecting help from helpers regardless of the self-other overlap.

-----------------------

[1] Our altruistic concerns factor did obtain two new items which seem more related to selfish concerns (see Table 2). However, all key analyses were checked using the original weights from Study 1 as well, and the same pattern of results held.

[2] These ² values are from a separate regression with N = 358.

These β values are from a separate regression with N = 358.

-----------------------

.17

.18

.37*** (.34)***

Higher-Cost Help

Rejected

Distressed

Altruistic Concerns

.28

.26

.22*** (.15)***

Lower-Cost䠠汥൰敒敪瑣摥഍䠍牯楲楦摥഍䤍慭敧䌠湯散湲൳䠍汥⁰敒敪瑣摥഍഍浉条湩⁥睏敆汥湩獧഍浉条湩⁥瑏敨鉲⁳敆汥湩獧഍浉条湩⁥湩传桴牥玒匠潨獥഍഍഍഍

Help

Rejected

Horrified

Image Concerns

Help Rejected

Imagine Own Feelings

Imagine Other’s Feelings

Imagine in Other’s Shoes

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download