A Replication of Solomon Asch’s Conformity Studies

Fool Me Once, Shame on You; Fool Me Twice, Conformity: A Replication of Solomon Asch’s Conformity Studies

Abstract

The number of studies have stressed that conformity in social groups is a desire to be accepted, to emulate the actions of those perceived as socially important. Research presented in this paper displays the findings that social conformity is still a viable reality in society. First, our research in social conformity is described by past studies that have gained popularity in psychological circles, mainly Asch’s and Milgram’s work on conformity and obedience. Second, the paper goes into the present research conducted at Cochise College studies. Lastly, our modified replication experiments indicate that people diverge from others even under conditions where conformity seems assured. Our current study of conformity provides strong evidence that participants, given a set of actions or questions, will conform to confederate actions when those confederates are considered part of one’s social group, i.e., a trusted associate. There are less conforming actions when there is only a random grouping of participants to study design. Overall, the evidence that supports our research that participants will conform to the social investigation was found to be accurate. More research needs to be conducted into conformity and to the social aspects of why it occurs, which would assist social researchers to understand the complexity of social convergence.

Fool Me Once, Shame on You; Fool Me Twice, Conformity: A Replication of Solomon Asch’s Conformity Studies

 The idea of conformity has been in the spotlight of psychological understanding since the seminal study of Solomon Asch in 1951 (Asch, 1956), where unwitting participants were statistically noted to conform to others groupthink over their own. Asch’s original work was an offshoot of earlier ideas and research on perception by Sherif (1935). Asch’s research has branched out to other avenues in Psychology. One such study was how children use nonverbal assessment to understand who is the more dominant figure in a social setting (Brey and Shutts, 2015). Wherein, Brey and Shutts (2015) argues that there are questions as to how these patterns emerge, both verbally and physically that even children begin to innately understand who if the authority figure one should turn towards and conform. The available evidence, though, reveals a far more complex and interesting set of dynamics: divergence is as pervasive as convergence, but its appearance is often unnoticed or minimized. Hodges, Meagher, Norton, McBain., & Sroubek, (2014) found that there was as much of a tendency in social grouping to conform as well as to nonconform, which they referred to as convergence and divergence (Hodges, et al., 2014). Although, there is not simply that we need to pay more attention to conformity, but that we need to appreciate a larger set of dynamics: social understanding is impossible to reduce to convergence or divergence. With these assumptions, it might seem important to show how groups of people are influence by others, what are the forms common to them? Divergence, or nonconformity, is sometimes seen as relatively unexciting in Psychological studies since it is a natural outcome with the independence and idealism of individual thinkers.

We chose to replicate a modified version of Asch’s Line Experiment (Asch, 1956), in which an unwitting participant would believe they were taking a College survey to asses their level of comprehension. As the original study is almost as well known as Milgram’s study on how far someone would go to shock a participant on the orders of someone perceived to be an authority figure (Milgram, 1963). For that reason that the original conformity study is well known (Asch. 1956), there were slight alternations in the set-up and delivery of the research experiment.

Literature Review

Taking the action of conformity to the economic perspective Nord (1969), exposed an underlying idea that nonconformity was an awkward position to be. Both as an economic perspective, as in not being able to have a winning business solution, as well as a personal position if at odds with a group action. It was relayed in the article how social conformity could be used to repay a debt in a way. That being that lending social support to an individual was seen as a form of social credit. This social credit had the effect of being repaid in kind when a similar social need arose. But individuals that had not provided some form of social “concern” to another were less likely to have any reciprocity to their dilemma. This version of social conformity was noted as well in how one was perceived, such as when a participant was instructed to act either more approvingly or acted in an unapproved manner. Of course, the more polite, or conforming to another a participant was, the better they were perceived (Nord, 1969).

Age as a Factor of Conformity

 Age can be considered a factor in how a respondent will answer a given question when placed in a position of opposing views. Walker and Andrade (1996), found through their studies with Australian participants, that the persons age played a large factor in rates of conformity. The conformity is believed to show that as someone age it creates a condition where the person is more grounded in their own perceptions. The younger the participant— the higher the rate of conformity. Much of this reaction stems from the belief the younger the participant, the less likely they can conceptualize the question. Hence the less they trust their views in preference to the group consensus, which is in conjunction with if the stimulus that is being judged is ambiguous. The more ambiguous the test question or visual display, the better the rates of conformity are produced. Conversely, with unambiguous target displays, or visuals, the lower the rates of conformity, as when the age of the participant is older (Walker & Andrade, 1996).

Realizing the Difference

 As age is likely a factor in rates of conformity, there is also the aspect the participant, even of a younger age will realize they agree to fit into a group consensus. Sun and Yu (2016) study on five and six-year-old students and their conformity rates as to what their “peers” viewed as attractive or less attractive faces. First, the students were seated in front of a computer and asked to rate the pictures generated on the screen as attractive (4, 5, 6) or less attractive (1, 2, 3) (Sun & Yu, 2016). After the first portion was over the children were brought back after a short while to retake the test in a modified form. This next stage of the experiment the student was placed back in front of a computer. A computer that generated faces of people that had already been rated by unseen “peers” on the same attractiveness scale of low (1, 2, 3) or high (4, 5, 6). The second iteration was to see if the children would conform to the unseen “peer” rating of the attractiveness of the picture. There was ample evidence that the children did alter their views on the pictures that were more in line with the peer ratings(Sun & Yu, 2016), which is in line with social conformity to a group consensus. The children did fall back to their own belief in private, outside the view of the peers, as to their original rating of the pictures. On subsequent days of testing the same participants still aligned their group views with the peer ratings, as opposed to their private views (Sun & Yu, 2016).

Knowing the Dominate Actor in Social Groupings

Brey and Shutts, (2015), conducted a study of nonverbal communication into perceptions of children as to who they viewed as the more dominate actor in a social situation. The role that body posture and tone of voice are nonverbally perceived in children gives clues as to certain aspects of how individuals socially conform to a group setting. Much of adult perceptions were molded in youth as to who one viewed as the dominant person in any interaction. This same nonverbal assessment appeared to develop from social cues learned at younger ages.

Self-Esteem and Intelligence in Conformity

 Self-esteem has a large impact on rates of social conformity as Rhodes and Wood (1992) delineated in their hypothesis that persons with higher levels of intelligence have a greater level of influenceability on lesser intelligent counterparts. Their premise followed that someone judged as more intelligent is better versed on multiple subjects. While at the same time those with higher intelligence are not as likely to be influenced by outside circumstances or exhortations (Rhodes & Wood, 1992). Likewise, it was shown that a message conveyed to someone with low self-esteem is much more likely to be accepted than someone with higher self-esteem. While also someone with low-self-esteem use more animated defenses, such as projection and regression, which permit threatening information to be expressed. So that recipients of low esteem yield more to influences than persons of high esteem (Rhodes & Wood, 1992).

Convergence and Divergence of Conformity

 Hodges, Meagher, Norton, McBain., & Sroubek, (2014) came across a novel counterpoint to Asch’s (Asch, 1956), original study. There seemed to be a body of participants that would fail to conform at any cost through many studies. Asch’s original study concluded after the 12 research trials that approximately 25% of participants would not conform in any manner (Asch, 1956). Hodges, et al. (2014), noted a tendency of some participants to be in a position of speaking-from-ignorance (SFI). A complex idea that some participants would “diverge” from any conformity to knowledgeable authority, in the belief they were consistent with a truthful position. Much of what Hodges, et al. (2014) proposed was there was a body of participants that would converge, i.e., conformed to a group consensus. Also that there was a body of participants that diverged, i.e., disagreed with a consensus view. Their findings pinned much to the theory that, “divergence is as pervasive as convergence in situations (e.g., imitation, mimicry, majority influence) where the emphasis has been almost solely on convergence[in conformity studies]” (Hodges, et al.,2014).

Relational Conditions

Russell (2014) went back to Milgram’s original studies in obedience to authority to find what connections lent much of its energy to get persons to be obedient and conform to a researcher’s commands, even seemingly painful ones. Russell found that Milgram began a study where he asked participants to bring a friend to act as the “learner,” while the original participant was the “teacher.” This was termed the Relationship Condition (RC). As in the original understanding of Milgram’s experiment the “learner” would not be harmed but was to act so, while the “teacher” was not let in on this position while giving painful shocks to the learner when they answered incorrectly. Russell found that Milgram’s RC experiment still had 15% of the teachers complete the full round of shocks on their “friend” (Russell, 2014).

Situational Forces

 Burger (2009) argues that situational forces of any research are affected when there is a change in surroundings that cannot be accounted for in a hypothesis. Which stated more clearly is that when a participant is placed in an unfamiliar setting with a perceived authority figure, a participant’s actions can alter from what they would consider they would normally decide as Burger (2009) asserted that Milgram believed that the cornerstone to obedience was more to do with how an authority figure was viewed. This played into our conformity research as well. The belief that confederates actions were correct, given that the confederate was a trusted counterpart. Social conformity has as much to do with obedience, as our Western culture places blind faith in certain authority figures, which could is extended to ones trusted associates (Burger, 2009).

Methods

 Researchers; Antje Rester, April Carroll and I obtained informed consent (see Figure 1) from all participants after asking verbally if they were at least 18 years of age, and if they wanted to participate in a college level survey. Researchers also maintained all confidential forms in separate locations. Meaning that the researchers had different folders and areas of safekeeping for both informed consent forms, as well as demographic questions (see Figure 2) from all participants.

Study 1

Participants

 In the first study, there were seven participants that stated they would like to participate. There were four females and three male participants. One participant classified themselves as a student and a staff member, as he worked in a dual role in the Learning Lab outside of his student status. The remaining six termed themselves as students of Cochise college.

Procedures

Our study randomly recruited participants from Cochise College Sierra Vista main campus by recruiting participants by a written survey on basic college knowledge (see Figures 3 and 4). Participants were offered water and doughnuts for their time as compensation for their participation. All participants rotated through a classroom with one confederate in attendance, which also rotated in and out of the class to provide legitimacy for his presence. The researchers placed an electronic speaker under a table where the participants sat. The signal was an electronic tone that sounded at 20-second intervals. After each tone, the confederate would stand up for a moment and then sit back down to continue with the survey. After each participant finished, they would give their survey to the researcher and be debriefed in a separate location away from the study classroom. The first research survey and design conducted on a March 6^th— in the morning.

Study 2

Participants

 During the second research day, there were a total of 12 participants. And three confederates. Of the 12 participants, five were female, and seven were male. Of those 12, ten defined themselves as students of Cochise College. One was a staff member at Cochise College and one participant identified as “other.” The mean age for the participants was the bracket of 18 to 24 years of age.

Procedures

Our second study also randomly recruited participants from Cochise College Sierra Vista main campus by recruiting participants by a written survey on basic college knowledge (see Figures 3 and 4). During this study, day participants were offered water, soda, candy, and some pastries as compensation for their participation. All participants rotated through a classroom with one or more confederates in attendance, which also rotated in and out of the class to provide legitimacy for their presence. The researchers placed an electronic speaker under a table where the participants sat. The signal was an electronic tone that sounded at 20-second intervals. After each tone, all confederate that was present in the room would stand up for a moment and then sit back down to continue with the survey. After each participant finished, they would give their survey to the researcher and were debriefed. For study 2 there were three confederates, and the study was conducted on March 25^th, again, in the morning.

Study 3

Participants

 There were three groups of 27 participants in total. Twenty-four were female, and two were male. All participants were students in various stages of Nursing or other related introductory medical fields. The mean age for this group was between 25 and 34 years of age. Confederates selected at random from the groups of volunteers that assembled for the survey. In group 1 there were 13 females, 4 were confederates. In group 2 there were four females, 2 were confederates. In group 3 there were three males and seven females, three females and one male were confederates.

Procedures

The final study research moved to the downtown campus of Cochise College which mainly deals with Nursing and related medical studies. The survey was altered to reflect medical knowledge and was verbally given by one of our researchers (see Figure 5). The confederates selected at random from the participants that gathered to take the survey. At which time they briefed before to answer incorrectly to all questions asked. Once back in the classroom, the confederates were seated at the front of the classroom so they could begin answering incorrectly and in succeeding order. Students that were unwitting participants would answer after most of the confederates by verbal responses. There were three groups of participants in this study. This study conducted on a Wednesday, April 3^rd, at lunchtime and participants were offered pizza, water, and soda for their time. Participants were debriefed and allowed to give feedback before leaving the classroom.

Results

Antje Rester and I hypothesized that participants would replicate a repetitive controlled conformity behavior. Overall our three separate studies generated 36% conformity in total. This percentage was in line with Asch’s original findings of his 12 study trials, which displayed at 36.8 % rate of “error,” or conformity to an incorrect answer (Asch, 1956, table 3, pg. 10). With our research, with study one (N = 7) with one confederate and study 2 (N = 12) with three confederates, there were no conformists (0% conformity). In the three groups of study 3 (N = 27) the results were as follows: in the first group (N =13) there were 4 confederates with 5 conformists (55% conformity), the second group (N =4) contained 2 confederates with 2 conformists (100% conformity), and the third group (N =10) 4 confederates yielded 6 conformists (100% conformity). When compiling the statistical data of all three studies (N = 46) using a Kruskal-Wallis test, we found of the 46 participants, 13 participants (36%) showed conformity behaviors. Of the 46 participants, ten participants were excluded from the total percentage as they were confederates. The Kruskal-Wallis H test statistic showed X2(2) = 7.591, p-value is .02247. The result was significant at p < .05. Our research rejected the null hypothesis, meaning participants conformed to a measured conformity behavior.

Discussion

Our research team sought to understand the rate of conformity in an alternate replication of Asch’s Conformity Experiment. Antje Rester and I hypothesized that participants would conform to confederate actions, as our research hypothesis. The results the research provided was that we should reject the null hypothesis, which was that there would be no change and no conformity during the research. This meaning that our research showed a significant level of participant conformity during our testing phase. This result was demonstrated by certain acts of conformity when participants were asked questions after a group of confederates answered incorrectly. Our original two study days used a physical form of confederate action –standing up during an electronic tone—which did not garner any observable conformity. Only the verbal question and answer during study three successfully produced conformity.

Successes

 Our study did not permit us to select participants randomly. Our research used random selection for confederates, and convenience sampling for the body of our unwitting participants. Throughout our three study iterations, we noted that the physical confederate action, standing, did not prove successful to create conformity. Differently, with a verbal medical Q & A, we collected data that showed significant conformity rates. This conformity result was dependent on what method of conformity test administered, whether physical or verbal. We demonstrated to ourselves that physical movements were not able to produce conformity, very likely due to lack of confederate numbers. While sufficient confederate numbers, as well as familiarity of confederates—known classmates— did prove successful in gaining conformity. These measures are taken to ensure the validity and reliability of the experiment was in keeping detailed notes as well as conducting after-research discussion to note our finding while fresh in mind. Our research conducted on days when we expected the student concentration on campus to be higher, which would ensure we had a larger pool of likely participants. Two members of the research team maintained the informed consent forms, as well as the completed participant surveys, separate, during all three study days, to assure that the anonymity of the participants was preserved.

Limitations

The greatest limitation found in the study was that of an adequate number of confederate participants. From the original study (Asch, 1956) there was a six to one ratio of confederates to participants. This level of confederates was unattainable for our research team to acquire during our test days. Mainly this operating issue was due to time constraints as well as a lack of participants that would be willing to act as confederates.

Another limitation was that depending on the conformity study that we employed, physical or verbal, created the difference if the researchers could elicit conforming actions or not. The physical test, standing up at the sound of an electronic tone, was limited in the amount of conformity seen due to a paper test given to participants. With the written survey the participants did not pay attention to the confederates actions of standing to the sound of an electronic tone. The lack of participant attention to their surroundings became apparent after the first trial was completed. As the researchers noted that every participant seemed to be engrossed only in their survey questions and rarely seemed attentive to any confederate actions.

Additionally, the effect of some form of incentive appeared to play a significant role in how many participants could be recruited at any location. Some form of food attracted most of our participant pool, with an offer of free pizza being the largest incentive noted.

Explanations of Results with Comparisons to Previous Research

Our study found similar results and conformity rates to that of Solomon Asch’s original study (Asch, 1956), where once we increased the number of confederates within a study group, we achieved significant levels of conformity. The question as to ‘why do some participants conform,’ and others do not became delineated in an article that spoke to the reasons behind conformity; and that of convergence (Hodges, et al. 2014). Hodges’ study noted a slight difference in why one conforms or converges with a group consensus, which seems to point to a statistical average in any grouping that contains some conformist and nonconformist. These findings did coincide in certain areas with the conclusions of previous research in conformity (Asch, 1956).

Implications

 Our research chose an altered replication of Asch’s Conformity Studies. Asch’s original study ranks in comparison to some of the most notable studies in Psychology, as concerning importance. Conformity in social grouping is essential to further understanding of the psychology of group dynamics as well as social hierarchies. The findings of our research proved that certain actions by confederate participants did not have a level of importance to gain conformity of participants. Also with additional confederates, and confederates that are known to unwitting participants, and likely respected by fellow participants— conformity results could be elicited.

Suggestions for Future Research

  Our results showed that given significant confederate numbers and actions can still create conformity in a group. Meaning that given enough inside participants who understood the direction the study was intended to elicit could have created more social conformity. More research needs to be conducted to show how social stigma, as well as social and psychological expectations, help create the conditions in participants to conform to improper reactions from confederate actors.

Our team conducted three studies at two different campus locations. Our results proved to be successful in showing that our research hypothesis, with two of the three team members, was the most accurate outcome—participants will conform to confederate actions. The findings of our research suggested that confederate actions do provoke conformity in certain settings. The research was significant enough to warrant further study into the effects of social conformity in groups which would have the implication that could better explain social groupings and their interactions.

References

• Asch, S. E. (1956). Studies of independence and conformity: I. A minority of one against a unanimous majority. Psychological Monographs, 70(9), 1-70. http://dx.doi.org.cochise.idm.oclc.org/10.1037/h0093718
• Brey, E., & Shutts, K. (2015). Children use nonverbal cues to make inferences about social power. Child Development, 86(1), 276-286. http://dx.doi.org.cochise.idm.oclc.org/10.1111/cdev.12334
• Burger, J. M. (2009). Replicating milgram: Would people still obey today? American Psychologist, 64(1), 1-11. http://dx.doi.org.cochise.idm.oclc.org/10.1037/a0010932
• Hodges, B. H., Meagher, B. R., Norton, D. J., McBain, R., & Sroubek, A. (2014). Speaking from ignorance: Not agreeing with others we believe are correct. Journal of Personality and Social Psychology, 106(2), 218-234. doi:10.1037/a0034662
• Milgram, S. (1963). Behavioral Study of obedience. The Journal of Abnormal and Social Psychology, 67(4), 371-378. http://dx.doi.org/10.1037/h0040525
• Nord, W. R. (1969). Social exchange theory: An integrative approach to social conformity. Psychological Bulletin, 71(3), 174-208. http://dx.doi.org.cochise.idm.oclc.org/10.1037/h0027032
• Rhodes, N., & Wood, W. (1992). Self-esteem and intelligence affect influenceability: The mediating role of message reception. Psychological Bulletin, 111(1), 156-171. http://dx.doi.org.cochise.idm.oclc.org/10.1037/0033-2909.111.1.156
• Russell, N. (2014). Stanley milgram’s obedience to authority “relationship” condition: Some methodological and theoretical implications. Social Sciences, 3(2), 194-214. http://dx.doi.org.cochise.idm.oclc.org/10.3390/socsci3020194
• Sherif, M. (1935). A study of some factors in perception. Archives of Psychology. 27th ed. p.187.
• Sun, S., & Yu, R. (2016). Social conformity persists at least one day in 6-year-old children. Scientific Reports (Nature Publisher Group), 6, 39588. http://dx.doi.org.cochise.idm.oclc.org/10.1038/srep39588
• Walker, M. B., & Andrade, M. G. (1996). Conformity in the asch task as a function of age. The Journal of Social Psychology, 136(3), 367-372. Retrieved from https://cochise.idm.oclc.org/login?url=https://search-proquest-com.cochise.idm.oclc.org/docview/78194082?accountid=7278

Figure 1: Informed consent

Figure 2: Demographic form

Figure 3: Test version 1a

Figure 4: Test version 1b

Figure 5: Medical Question and Answers