Abstract
The effects of interpersonal commonalities on compliance to requests were measured among 15 female undergraduates. Subjects were put in four different conditions in which they were to fill out a questionnaire, answering intermingling questions based on their compliance to a hard and easy task for someone who they have a commonality with (an in-group member), and for someone who they did not (an out-group member). Subjects were to rate their level of compliance on a scale of 1-5. Consistent with our hypothesis, main effects of both task difficulty and group association were found, showing a greater level of compliance for easy requests versus hard, and for in-group members opposed to out-group. Social psychological reasons for the results are discussed.
Interpersonal Commonalities and their effect on Compliance to Requests
Every morning we wake up, get ready, and carry out our daily routine. We subject ourselves to many incoming stimuli, most of which is ignored. We process what is immediately relevant to our situation, but with so much coming in, there is no way we can process everything, there is simply not enough time. Suppose a student on his way to class to take a midterm, sees a girl walking a fast pace, looking frantically over her notes. There may be a large number of reasons to explain her behavior, perhaps it is the only time she has to study? On the other hand, maybe she is simply looking over a paper for a friend. Since most of us do not have the time to come up with all the possible reasons, we take a “cognitively efficient approach, and rely on well learned scripts” called heuristics. (Burger, Messian, Patel, Prado & Anderson, 2004) As a college student, he will most likely infer that she has a midterm and is putting in some last minute study time. What he did was notice a commonality between himself and her (they were both students at the same university), and came to a conclusion, probably without even thinking about it too much. Suppose his inference was correct, and as he continued on his way, she approached him. He then recognizes her as one of his classmates. She asks him if she can borrow his notes from a lecture that she missed. What does he say? According to research by Burger et al., (2004) “heuristic processing can lead to an increase in compliance when salient cues indicate that this is the kind of person we usually say yes to” (Burger et al., 2004). This means that because they both shared something in common, he is more likely to comply with her request rather than if she were a stranger. Burger et al., (2004) researched just that, he and his partners conducted three separate experiments where a confederate who shared a common trait with the subject (a birthday, fingerprint, or a first name) interacted with the subject. The results found that levels of compliance increased when the subjects felt that the commonality was significant, however compliance did not change if the shared trait with the requester was prevalent.
For most of us, helping someone else feels “good” inside, a feeling of euphoria from aiding someone other than ourselves, but this feeling of “good” is from knowing that we unconditionally helped someone in need. What if we were forced to help? Or more likely, pressured to help? Well according to findings by Williamson & Clark (1992) “affective reactions to providing help are moderated by the type of relationship that helpers desire with the recipient” (Williamson & Clark, 1992). What Williamson and Clark conducted in 1992 was a study on subjects who were led to desire a close intimate relationship, or a distant acquaintance type relationship, with a hypothetical subject. The subjects in this study were then either required to help the hypothetical subject, or chose to help; their reactions were then examined and measured for positive or negative affect. What Williamson & Clark (1992) found was that positive affect went up when the subjects helped someone who they desired a close relationship with, regardless of whether or not they were required to. These findings by Williamson and Clark lead to another interesting study by Burger, Soroka, Gonzago, Murphy, & Somervell (2001) who looked at the effects of fleeting attraction on compliance to a request. Burger et el., (2001) found that increasing the temporary feeling of attraction leads to an increase in compliance because we are more likely to help those who are emotionally close to us, or those who are similar to us. In line with Williamson & Clark’s (1992) results, we help others, or in this case, comply with their requests because it makes us feel good. Our study tests to what length a person will comply, factoring in the rewards (good feelings) and the costs (gravity of the favor/request).
Past research has found evidence for the effects of closeness or similarities among people on compliance. What we wanted to measure was if the compliance for a difficult request would be the same for an easy request if the subject felt a close relationship with the requester. We predicted, because of similar research done in the past, that compliance will be significantly higher when the requester is someone who the subject shared a commonality with (who we will call “in-group” members). We also predicted that level of compliance be higher for an easy request versus a hard request. Looking at the levels of task difficulty (easy and hard), we predicted that compliance for an easy request would be slightly higher for in-group members than out-group, however, compliance to a hard request for an out-group member would be far less than that of an in-group member, thus we predicted an interaction where compliance to the difficulty of the task depended on group association. The participants in our study were all part of the same lab section seven weeks prior to the experiment, so they shared some level of closeness. However, within the lab section, they were broken into groups of three for an experiment six weeks prior to testing, so they spent much more time with members of their experiment groups (in-group members) leading to a closer relationship among them. The in-group member status within the subjects, that is, the two people who are members of their experiment group were compared with two out-group members (other classmates not in their group) for compliance to a request.
Method
Participants
Fifteen female undergraduate students in a psychology research methods course at the University of California, Los Angeles were tested in this study. The subjects were in the same lab section for the research methods course for nine weeks prior to the experiment, and they were in their unique lab experiment groups for six weeks prior to testing. The subjects were not paid for their participation; rather it was made mandatory for full completion of the course.
Design
The study was conducted using a 2x2 within subjects design. Subjects were numbered from 1 to 15 based on their seating in the class. Each subject was assigned to evaluate four people: two in-group members (their experiment groups) and two out-group members (other classmates).
Materials and Apparatus
A random number table was used to pair the subjects with the out-group members. A survey with eight requests, four easy and four hard were given to each subject along with a piece of black construction paper to immediately cover their responses. There were 15 different versions of this survey, all asking the same requests, only the order and sequence of the requests varied. The questions were balanced for order and sequence effects using a balanced Latin square within a Latin square design (appendix B). The easy and hard categories were agreed upon by three independent judges who had a high and consistent agreement (95%) on task difficulty. Each request was followed by a compliance scale of 1-5. The questions were worded in a clear and concise manner; the subjects were given a hypothetical situation involving a request from one of their assigned people, and on the scale, rate how compliant they would be to the request. The eight compliance requests on the survey which the subjects had to answer were split between four easy and four hard requests. An “easy” request was defined as a request which one would likely encounter on a daily basis, and would be complied with more than 50% of the time, (e.g., lending a small amount of change). A “hard” request was defined as a request which one would not encounter very often, and in which compliance would be less likely (e.g., lending twenty dollars). Compliance was measured on an interval scale of 1 to 5 following each question, 5 being completely compliant, 3 being indecisive, and 1 being completely noncompliant Each question had a blank spot (prior to handing them out) for the subjects to fill in with the name of the person being evaluated. A copy of the survey can be found in appendix A for details of the compliance scale and the specific requests presented.
Procedure
As the experiment began, the subjects were seated. They were asked to make sure that they were seated respectively with their experiment (in-group) group member so that each of the five experiment groups would be seated together. Counter-clockwise from the last table, the subjects were numbered from 1-15. Once they had their numbers assigned, the middle person seated in each in-group was asked to switch seats with the middle person of the group counter clockwise to them; this was so no in-group members were seated next to each other while completing the survey. Once the seating was arranged, a random number table was used to assign two out-group members to each subject based on their assigned numbers. Precautions were taken so that none of the assigned out-group members were from the subject’s in-group, and none of the subjects were assigned the same out-group member twice. The questions were randomized using balanced Latin squares (Appendix B) so that each subject received the requests in a different order, thus eliminating all sequence and order affects. This design allowed all subjects to be evaluated four times (twice by two different out-group members and twice by their in-group members). Once all of the subjects were assigned to out-group members, they were told that they were going to take a survey. Before receiving the survey, they were instructed to put their assigned number at the top, answer the questions one at a time in order, and cover their responses immediately after marking it down using the black construction paper. Once they were finished, they were to cover their entire survey with the black construction paper. Before handing out the surveys, the names of the four people being evaluated were filled in the blank spots for each question by the subjects. The name of each subject was written on a whiteboard next to their corresponding number. The subjects were then read aloud their two out group members by number, and were to write down the numbers at the top left of their paper. They were then told to alternate each page of questions by answering first for “in-group” if their assigned number was even, and for “out-group” if their assigned number was odd. The subjects were told neither what the survey was about, nor what it was testing. The surveys were then handed out face down in the same counter-clockwise design (based on seating position) and once all subjects had a survey they were told to begin. When all the subjects were finished, the surveys were collected in an orderly fashion with the construction paper on top, but also briefly looked over by the experimenter to make sure that all of the questions were answered, and that the subjects had put their number on top of the page. Once all surveys were collected, subjects were allowed to ask questions regarding the relevance of the study, and any concerns they may have had.
Results
Figure 1 displays the average level of compliance when the subjects were presented with easy versus hard requests compared with who they did the tasks for. The patterns in figure 1 show that in general, easy requests were complied with more than hard requests, and requests made by in-group members (shown as “group”) showed a higher rate of compliance than for the out-group (shown as “person“). The level of compliance based on in-group/out-group members however, did not seem to depend on the difficulty of the task, the two variables seemed to work independently in their own way.
To test the effects of compliance, a 2x2 analysis of variance within subjects was used to analyze the data. The test revealed a significant main effect of group association (in-group/out-group), hence the average level of compliance for in-group members was significantly higher (M= 4.3, SD = .85496) than for out-group members (M= 3.633, SD= .76064), regardless of task difficulty, F(1,14) = 9.492, MSE = .702, p = .008. A significant main effect was also found on the type of request, with easy requests showing a significantly higher level of compliance (M= 4.65, SD= .67788) than hard requests (M= 3.2833, SD= 1.27055), regardless of group association, F(1, 14) = 87.487, MSE = .320, p = .000. The interaction between type of request and group association however was not so significant, F(1,14) = 2.441, MSE = .335, p = .140.
In order to compare the individual means within each condition, four pair wise comparisons (with alpha level of .05 maintained using a Bonferroni correction) were conducted. When presented with easy requests, the average level of compliance was significantly higher when processed with in-group status (M = 4.8667, SD = .22887) as opposed to out-group (M = 4.4333, SD = .65101), t(14) = 2.476, p< .027. In addition, when presented with hard requests, the average level of compliance was also significantly higher for in-group status (M = 3.7333, SD = .84233) than for the out-group (M = 2.8333, SD = 1.06346), t(14) = 2.743, p< .016.
Furthermore, when looking at in-group status, the average level of compliance for easy requests (M = 4.8667, SD = .22887) were found to be significantly higher than for hard requests (M = 3.7333, SD = .84233), t(14) = -5.134, p< .000. Additionally, for out-group status, the average level of compliance was found to also be significantly higher for easy requests (M = 4.4333, SD =.16809), than for hard requests (M =2.8333, SD = 1.06346), t(14) = -8.147, p< .000.
Discussion
In our study we found that interpersonal commonalities do in fact result in increased levels of compliance, consistent with past studies. In addition easy requests lead to greater levels of compliance than hard requests, both consistent with our predictions. The interaction between task difficulty and group association however was not so significant, contrary to our predictions. The difficulty of the request was not dependent on whether or not it was for an in-group or out-group member. In general, subjects preferred complying with easy tasks more than hard, and with in-group members more that out-group.
On a general consensus, our findings were similar to the supporting articles by Burger et al., (2001, 2004) where commonalities between people lead to an increase in compliance. We like people who are similar to us, thus from Williamson & Clark’s (1992) study, for people who we want a communal relationship with, we feel good when helping them. From our study, we found that the opportunity cost of feeling good by helping our “in-group” members outweighs the cost of having to do them a favor, which is a possible reason as to why compliance levels are higher for in-group than for out-group members, where the cost of having to do the favor for someone probably outweighes the philanthropic feeling of good.
Our study augmented and strengthened the existing theories of commonalities on compliance to requests, which in turn leads to increased knowledge in the world of research. Criterion, construct, and face validity are all strong in our experiment, as well as internal validity however, because our subjects were all female undergraduate students, external validity was weakened, as our subject pool only accounted for a small percentage of the population. This however, should not be of too much concern as our results were consistent with the findings of Burger et al., (2001, 2004), so it is safe to say that our sample of subjects tested relatively similar to people in general.
Research on compliance and persuasion are big in the field of social psychology, and there is always room to add to, and improve on current research. Our study looked at compliance requests among females, perhaps in the interest of research, one could study the effects of compliance on males to females and vice versa. Furthermore, for future research, one could also test the effects of the “foot in the door” phenomenon on compliance, which is the theory that once a simple request is asked, people are more likely to comply with a hard request than if no simple request was asked to begin with. Because psychology is a probable science, there is an endless amount of research that can be done of any one topic; therefore, although it is a good practice to stand by the results, one should also be open to new and possibly counter-evident research.
References
1) Burger, JM., Messian, N., Patel, S., Prado, AD., Anderson, C. (2004). What a Coincidence! The effects of incidental similarity on Compliance.Personality & social psychology bulletin, 30(1), 35-43.
2) Burger, JM., Soroka S., Gonzago, K., Murphy, E., Somervell, E. (2001). The effect of fleeting attraction on compliance to requests.Personality & social psychology bulletin, 27(12),
1578-1586.
3) Williamson, GM., & Clark, MS. (1992). Impact of desired relationship type on affective reactions to choosing and being required to help.Society for Personalityand Social Psychology, 18(1), 10-18
